Stimulation of Tree Defenses by Ophiostomatoid Fungi Can Explain Attack Success of Bark Beetles on Conifers François Lieutier, Annie Yart, Aurélien Salle

Home , Grosmannia, Grosmannia clavigera, Ophiostoma

Stimulation of tree defenses by Ophiostomatoid fungi can explain attack success of bark beetles on conifers François Lieutier, Annie Yart, Aurélien Salle

To cite this version:

François Lieutier, Annie Yart, Aurélien Salle. Stimulation of tree defenses by Ophiostomatoid fungi can explain attack success of bark beetles on conifers. Annals of Forest Science, Springer Nature (since 2011)/EDP Science (until 2010), 2009, 66 (8), 10.1051/forest/2009066. hal-00883599

HAL Id: hal-00883599 https://hal.archives-ouvertes.fr/hal-00883599 Submitted on 1 Jan 2009

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Ann. For. Sci. 66 (2009) 801 Available online at: c INRA, EDP Sciences, 2009 www.afs-journal.org DOI: 10.1051/forest/2009066 Review article

Stimulation of tree defenses by Ophiostomatoid fungi can explain attack success of bark beetles on conifers

François Lieutier1*, Annie Yart2, Aurélien Salle1

1 Université d’Orléans, Laboratoire de Biologie des Ligneux et des Grandes Cultures UPRES EA 1207, rue de Chartres, BP 6759, 45067 Orléans Cedex 2, France 2 INRA, Unité de Zoologie Forestière, 2163 avenue de la Pomme de Pin, CS 40001, Ardon 45075 Orléans Cedex 2, France

(Received 24 November 2008; revised version 23 June 2009; accepted 7 July 2009)

Keywords: Abstract pathogen / • Our aim is to present why the hypothesis, that Ophiostomatoid fungi play an important role in the forest tree / establishment of most bark beetle species on living conifers, is valuable. physiological stress / • After summarizing knowledge about the relationships of bark beetles with conifers and fungi, we blue stain fungi / conclude that controversy results from misinterpretations when using fungal pathogenicity to demon- host resistance strate the role of Ophiostomatoid fungi in beetle establishment on host trees. • We demonstrate that fungal pathogenicity is not the right parameter to appreciate the role of fungus in beetle establishment on host trees. We argue that artificial low density inoculations that allow the appreciation of fungus ability to stimulate tree defenses and thus to help beetles in overcoming tree resistance must be used in complement to mass inoculations. In both cases, results must be expressed in terms of tree defense stimulation rather than in terms of tree killing. (i) Fungal species stimulating tree defenses are generally not those that grow the best in the sapwood. (ii) We argue that beetle development in the phloem, fungal invasion of the sapwood and phloem, and tree death, occur after tree defenses are exhausted, and that any fungus present in the beetle gallery could thus potentially invade the sapwood after defense exhaustion. • We conclude that stimulation of the tree defense reactions in both the phloem and the superficial sapwood is a real benefit brought by fungi to the beetles during the first phase of establishment (overcoming tree resistance). • Considering the origin of the bark beetle fungus associations attacking living trees and their general functioning based on stimulation of tree defenses, we develop three hypotheses: (i) any beetle species would be helped in its establishment in a given tree species by developing an association, even loosely, with a fungus species belonging to the Ophiostomatoid flora of that tree species; (ii) the necessity of a considerably low level of tree resistance for fungus extension into the tree is the selection pressure that has led fungi to develop their intrinsic ability to stimulate tree defenses, through their ability to grow into the phloem. This association can be completed by antagonistic fungal species controlling extension of the previous fungal species in the tree tissues; (iii) Beetle species using the strategy of overcoming tree resistance are associated with a fungal complex, of which species could assume three roles regarding relationships between beetles and trees: 1- to stimulate tree defenses in the phloem and superficial sapwood, 2- to grow into the sapwood after tree resistance is overcome, and 3- to control phloem extension of the first other two categories. Bringing nutrients to the beetle progeny can be a fourth role. • We propose that bark beetle – Ophiostomatoid associations can be categorized, based on associations’ frequency and complexity while taking into account beetle aggressiveness. We show that a close correspondence exists between beetles’ aggressiveness and the ability of their main associated fungal species to stimulate the defenses of their host tree. • We conclude with suggesting that most sapwood invading fungi might be “cheaters” which have taken advantage of the efficiency of the relationship between beetles and fungi that stimulate tree defenses.

* Corresponding author: [email protected]

Article published by EDP Sciences Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

Mots-clés : Résumé – La stimulation des défenses de l’arbre par les champignons Ophiostomatoïdes peut pathogène / expliquer le succès des attaques de Scolytes sur conifères. arbre forestier / • Notre objectif est de présenter les raisons de la validité de l’hypothèse selon laquelle les champi- stress physiologique / gnons jouent un rôle important dans l’installation de la plupart des espèces de Scolytes sur conifères champignon du bleuissement / vivants. résistance de l’arbre • Après avoir résumé les connaissances sur les relations des Scolytes avec les conifères et les champignons, nous concluons que la controverse résulte d’interprétations erronées lorsque l’on utilise le pouvoir pathogène des champignons pour démontrer le rôle des Ophiostomatoïdes dans l’installation des insectes sur les arbres hôtes. • Nous démontrons que le pouvoir pathogène n’est pas le paramètre correct pour apprécier le rôle du champignon dans l’installation des Scolytes sur les arbres hôtes. Nous soutenons que des inoculations artificielles à faible densité, qui permettent d’apprécier la capacité du champignon à stimuler les dé- fenses de l’arbre et à ainsi aider le Scolyte à surmonter la résistance de celui-ci, doivent être utilisées en complément des inoculations massives. Dans les deux cas, les résultats doivent être exprimés en termes de stimulation des défenses de l’arbre plutôt qu’en termes de mortalité de l’arbre. (i) les espèces de champignons qui stimulent les défenses de l’arbre ne sont généralement pas celles qui présentent la meilleure croissance dans l’aubier. (ii) nous soutenons que le développement de l’insecte dans le phloème, l’invasion de l’aubier et du phloème par le champignon, et la mort de l’arbre, interviennent après épuisement des défenses de l’arbre, et que tout champignon présent dans les galeries de l’insecte pourrait donc potentiel- lement envahir l’aubier après épuisement de ces défenses. • Nous concluons que la stimulation des réactions de défense de l’arbre à la fois dans le phloème et l’aubier superficiel représente un bénéfice réel apporté par les champignons aux Scolytes pendant la première phase de leur installation (surmonter la résistance de l’arbre). • En ce qui concerne l’origine des associations Scolytes – champignons attaquant les arbres vivants et considérant leur fonctionnement général basé sur une stimulation des défenses de l’arbre, nous développons trois hypothèses : (i) toute espèce de Scolyte serait aidée dans son installation sur une espèce d’arbre donnée en dé- veloppant une association, même lâche, avec une espèce de champignon appartenant à la flore Ophiostomatoïde de cette espèce d’arbre ; (ii) la nécessité d’un très faible niveau de résistance de l’arbre pour autoriser l’extension fongique dans le végétal est la pression de sélection qui a conduit les champignons à développer leur capacité intrinsèque de stimulation des défenses de l’arbre, à travers leur capacité à croître dans le phloème. Cette association peut être complétée par des espèces fongiques antagonistes contrôlant l’extension des espèces précédentes dans les tissus de l’arbre ; (iii) les espèces de Scolytes utilisant la stratégie de surmonter la résistance de l’arbre sont associées à un complexe fongique dont les espèces assurent trois fonctions eu égard aux relations entre les Scolytes et les arbres : 1– stimuler les défenses de l’arbre dans le phloème et l’aubier superficiel, 2– croître dans l’aubier après que la résistance de l’arbre ait été vaincue, and 3– contrôler l’extension des deux catégories précédentes dans le phloème. L’apport de nutriments à la progéniture du Scolyte peut représenter une quatrième fonction. • Nous proposons que les associations Scolytes – Ophiostomatoïdes puissent être classées, en se basant sur la fréquence et la complexité de l’association et en prenant en compte l’agressivité de l’insecte. Nous montrons qu’il existe une étroite correspondance entre l’agressivité des insectes et la capacité de leur principale espèce fongique associée à stimuler les défenses de l’arbre hôte. • Nous concluons en suggérant que la plupart des espèces de champignons envahissant l’aubier pour- raient être des “tricheurs” qui ont profité de l’efficacité des relations entre les Scolytes et les espèces fongiques stimulatrices des défenses de l’arbre.

1. INTRODUCTION Whitney, 1982, among others). All species bore galleries in their host, where larval development and most often adult mat- Bark beetles (Coleoptera: Scolytinae, Curculionidae) repre- uration take place. However, with the exception of the species sent a diversiﬁed group of insects gathering a little more than involved in Dutch elm disease, bark beetles of hardwood trees 6 000 described species worldwide and at least 225 genera do not generally cause much damage. Conifer bark beetles (Knizek and Beaver, 2004), specialized in exploiting woody on the contrary are directly responsible for frequent exten- plants (hardwood trees and shrubs, and conifers) and exhibit- sive and dramatic damage to forests, making them the most ing a high diversity of associations with micro-organisms important forest pests in the temperate zones (Grégoire and (Harrington, 2005; Kirisits, 2004; Paine et al., 1997;Six,2003; Evans, 2004). Understanding how their attacks succeed is thus

801p2 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801 essential, and their relationships with host trees have been doc- trees. They usually develop on logs and dead trees, eventu- umented in a variety of situations. Moreover, their frequent ally killed by species of the other groups, but a few species relationships with fungi have also stirred up many studies aim- are able to attack very weakened living trees. Although this ing at determining the role of these micro-organisms in the in- latter group represents around 95% of the total bark beetles teractions between trees and bark beetles. species in the temperate zones and plays an important ecolog- We believe that fungi play an important role in bark beetle ical role, few studies have been devoted to them because of establishment on conifers, a hypothesis first proposed by Reid their very low economical interest outside of those vectoring et al. (1967) and Berryman (1972) that has raised a lot of con- tree pathogens. The near-obligate and the facultative parasites troversy, and our objective in this paper is to present and jus- together make up the quasi-totality of species directly causing tify our positioning and our views regarding this hypothesis. damage to trees. We re-examine the relationships between trees and bark bee- Besides the saprophytic species, all groups must face a cer- tle – associated fungi, to propose that fungi effectively play a tain level of tree resistance before succeeding in establishing role in beetle establishment on trees, based on the stimulation in their host. Conifer resistance to bark beetle attacks involves of tree defenses and not on tree killing. We then propose hy- multiple passive or active defense mechanisms which over- potheses to explain why and how the beetle-fungi relationships lap and complete each other (Berryman, 1972; Christiansen interfere in this context. We finally propose different types of et al., 1987; Franceschi et al., 2005; Lieutier, 2002; 2004; beetle-fungus associations based on our hypotheses. Paine et al., 1997). Among these mechanisms, the hypersen- Before proceeding however, we will first summarize what is sitive reaction which develops in response to beetle attacks known about relationships between bark beetles and conifers has been demonstrated to play an essential role in the resis- on the one hand, and between bark beetles and their associated tance of many conifer species to many bark beetle species micro-organisms on the other. (Christiansen et al., 1987; Lieutier, 2002; Paine et al., 1997; Raffa, 1991;Raffa and Berryman, 1982a; 1982b; Safranyik et al., 1975). It develops around each point of attack in both the 2. RELATIONSHIPS OF BARK BEETLES phloem and the superficial sapwood and is visible as a longi- WITH CONIFERS AND FUNGI tudinal elliptical resin impregnated zone associated with more or less extended cell necrosis (Berryman, 1972;Reidetal., 1967). Such a zone is considerably enriched with terpenes and 2.1. Bark beetles and conifers neosynthesized phenolic compounds (Bois and Lieutier, 1997; Brignolas et al., 1995; Delorme and Lieutier, 1990; Långström Although some conifer bark beetles species develop inside et al., 1992; Lieutier et al., 1991a; Paine et al., 1987;Raffaand the sapwood, the vast majority are phloem borers. Four groups Berryman, 1982a; Russell and Berryman, 1976) contributing of conifer bark beetle species can be distinguished based on to stop the aggressors. Resin flow released by a section of resin the type of relationship they maintain with their hosts. The ducts can constitute another important resistance mechanism, “true parasites” can develop and reproduce on healthy trees at least for beetles of which the females bore transversal egg without killing them and this group seems to contain only two galleries (Berryman, 1972; Lieutier, 2002). species, which both live on spruces, the European Dendroc- Depending on how they cope with tree defense mechanisms tonus micans and the North American Dendroctonus punc- when they establish on their host, it has been suggested that tatus. The “near-obligate parasites” (Raffaetal.,1993)are bark beetle species attacking living conifers fit in with one or regular tree-killers able to kill healthy trees over large areas the other of two strategies (Lieutier, 1992; 2002). The near- during periods of outbreaks and which survive on subdued or obligate and facultative parasites would use the strategy of dying trees when their population is at a low level. The group overcoming host resistance (or exhausting host defenses). The is composed of a small number of species, amongst which are insects stimulate the syntheses of defensive compounds and Dendroctonus ponderosae, Dendroctonus brevicomis, Den- thus increase the energy demand until the tree becomes un- droctonus frontalis, attacking North American pines and Ips able to mobilize energy rapidly enough (Christiansen et al., typographus attacking spruces in Europe. Following its mas- 1987). At this moment, the tree’s defenses are exhausted and sive outbreaks in extended areas of spruces in North-West beetle establishment succeeds. The stimulation of these syn- America in the 1990’s, Dendroctonus rufipennis can also be theses is achieved basically through beetle aggregation on the considered a near-obligate parasite. The “facultative parasites” target trees (Berryman, 1972, 1976;Raffa, 2001;Raffaand (Raffaetal.,1993) usually colonize fallen or weakened trees Berryman, 1983a; Wood, 1982), which multiplies the points but are occasionally able to colonize healthy trees and to cause of aggression and thus increases the mechanical stress induc- significant damage, especially when there are circumstances ing and stimulating the tree’s hypersensitive response (Lieutier favorable to a considerable increase of their populations on et al., 1995). Above a critical threshold of attack density weakened trees (important storm felling for example). Species (Berryman, 1976;Raffa, 2001;Raffa and Berryman, 1983a; such as Ips pini on pines in North America, or Ips acumina- Safranyik et al., 1975; Thalenhorst, 1958), tree resistance is tus and Ips sexdentatus on pines and Pityogenes chalcogra- depleted to such a low level that it is not able to stop aggres- phus on spruces in Europe, belong to this group. The “sapro- sors anymore, finally resulting in beetle establishment and tree phyte” group (Raffaetal.,1993) comprises the large majority death. If it allows limiting interspecific competition because a of species which never colonize and reproduce on healthy low number of beetle species are able to attack living trees,

801p3 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al. such a strategy however can lead to intraspecific competition associated with glandular cells. Transportation within the my- (Raffa, 2001;Raffaetal.,1993). The opposite is true for the cangia would also give fungi a protection from UV light and strategy of avoiding host defenses, which would be used by desiccation (Klepzig and Six, 2004;Six,2003). the true parasites. Beetles behave in order to minimize the The existence of mycangia, which has arisen several times development of the tree’s hypersensitive reaction. They thus independently in the Scolytinae (Six, 2003), demonstrates that do not need to aggregate and they attack their hosts individu- bark beetles also have real benefits from the association. The ally. They also bore transversal galleries to avoid stimulating nature of the benefits however is not always very clear. Their the hypersensitive reaction too much, but this behavior causes associations with Ophiostomatoid fungi seem to bring bark the cutting of numerous resin ducts, which implies high bee- beetles diverse benefits depending on both beetle and fungal tle tolerance to constitutive resin (Lieutier, 2002). In such a species (Harrington, 2005; Paine et al., 1997;Six,2003). Nu- strategy which allows avoiding both interspecific and intraspe- trition is certainly important since, except for the true parasitic cific competition, tree resistance is not depleted and the tree species, bark beetles larvae feed on dying or dead trees, that is stays alive after attack success and during beetle brood devel- on nutritionally poor substrates. Considering the near-obligate opment. and facultative parasitic species, it is thought that fungi can im- prove larval diet through modifying the substrate or providing complementary nutrients such as vitamins, proteins or sterols 2.2. Bark beetles and their associated micro-organisms (Klepzig and Six, 2004;Six,2003). Evidence for a role of mycangial basidiomycetes in helping D. frontalis to meet its ni- Among the numerous species of micro-organisms (nema- trogen needs has been found (Ayres et al., 2000). Similarly, the todes, fungi, yeasts, bacteria) described as associated with presence of the Ophiostomatoid species Grosmannia clavigera bark beetles, fungi have been the subject of most studies (Six, and Ophiostoma montium can increase the nitrogen levels in 2003; Whitney, 1982). Associations with Basidiomycetes, the phloem of trees attacked by D. ponderosae by 40%, proba- mainly from the genus Entomocorticium, have been de- bly through their ability to restitute nitrogen from the sapwood scribed in D. ponderosae, Dendroctonus jeffreyi, D. brevi- to the phloem (Bleiker and Six, 2007), and all larval stages and comis, D. frontalis, I. typographus and Ips avulsus (refer- the adults of the insect can feed on the fungi (Adams and Six, ences in Six, 2003; and Kirisits, 2004) but most bark beetle 2007). More generally, mycophagy on Ophiostomatoid fungi, associated fungi are Ascomycetes, which have indeed been or improvement of larvae development in relation to the pres- isolated from practically all species where they have been ence of such fungal species, has been reported on several oc- looked for. Bark beetle associated Ascomycetes mostly be- casions (references in Six, 2003; Six and Klepzig, 2004;and long to the genera Ophiostoma, Ceratocystis and Ceratocys- Harrington, 2005), and Six and Klepzig (2004) have specu- tiopsis and their related asexual form Leptographium.These lated that all Dendroctonus species, except the true parasitic genera form, with few other genera, the morphologically ho- ones, would feed on fungi. According to Harrington (2005) mogenous group of Ophiostomatoid fungi (Kirisits, 2004;Six, however, fungal feeding does not seem obligatory for the com- 2003;Wingfieldetal.,1993), which are often referred to as pletion of the life cycle of bark beetles, and fungi would only “blue-stain”, “black-stain” or “sap-stain” fungi, because the supplement larval diet, leading to shorter larval galleries and melanized hyphae of most species give a bluish-grey color thus reducing both intraspecific and interspecific competition. to the sapwood they colonize, mostly in conifers (Harrington, Based on their nutritional relations with fungi and the local- 2005; Kirisits, 2004). ization of their galleries, bark beetles species have been tra- Functional relations between bark beetles and Ophiostom- ditionally divided into three groups, the xylomycetophagous atoid fungi correspond to various degrees of interactions in- species which bore galleries in the sapwood and feed exclu- cluding antagonism, commensalism and mutualism, faculta- sively on symbiotic fungi, the phloeomycetophagous species tive or obligatory (Klepzig and Six, 2004;Raffa and Klepzig, whose galleries are located in the phloem and which feed on 1992;Six,2003; Six and Klepzig, 2004). Considering bark both phloem and fungi, and the phloeophagous species which beetle species attacking living trees, the most important benefit feed only on the phloem (Francke-Grosmann, 1967; Kirisits, that Ophiostomatoid fungi usually gain from the associations 2004). is transport by the insect, on which they completely depend Apart from the nutritional aspect, another important benefit for their dissemination (Klepzig and Six, 2004;Six,2003). that conifer bark beetles could gain from the association would Facilitation in entering the available hosts through wounding be the help brought by Ophiostomatoid fungi to beetle species made by the beetles and limited interspecific competition is attacking living trees, in the success of their attacks and estab- also an advantage for fungi associated with beetles attacking lishment on trees. The reality of this benefit is the subject of living trees, compared to fungi developing on dead substrates. the present paper. Physical relationships are also very diverse. Fungal spores can be carried by beetles externally on the cuticle or inside more 2.3. Controversies or less complex mycangia (Harrington, 2005;Six,2003;Six and Klepzig, 2004) which, depending on beetle species and Following the pioneer work of Reid et al. (1967)andthe according to the classification of Six (2003), can be simple cu- first model of conifer-bark beetle-fungus interactions proposed ticular pits, pocket or tube sacs of varying depth in the integu- by Berryman (1972), blue stain fungi have been said to be re- ment, or setal brush, each of these structures being eventually sponsible for the development of the hypersensitive reaction,

801p4 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801 and to be essential for beetle establishment on conifers through 3. RELATIONSHIPS BETWEEN TREES AND BARK helping to overcome host resistance and killing the tree, both BEETLE – ASSOCIATED FUNGI AND REALITY beetles and fungi contributing simultaneously to tree death OF THE ROLE OF THE FUNGI IN BEETLE (Berryman, 1972). Establishment is said to begin (i.e. bee- ESTABLISHMENT ON TREES tles bore galleries and begin to lay eggs) when host resistance stops and death is assured (Berryman, 1982; Coulson, 3.1. Definition and quantification of beetle 1979; Wood, 1982). However, these early propositions have aggressiveness and fungal pathogenicity often been misinterpreted as it is necessary for the tree to be already killed by fungi for beetle establishment to begin. According to Raffa and Smalley (1988), aggressiveness is Then, studies developed to prove the essential role of associ- “used to denote the relative degree of vigor characterizing ated fungi in beetle establishment on trees aimed at demon- trees that can be colonized by a Scolytid species”. Thus ag- strating the pathogenicity of these fungi for the beetle host gressiveness is synonymous with the ability of the beetle to trees, mainly by using artificial mass inoculations supposed to become established on a living tree. Consequently, bark bee- mimic bark beetle attacks (Christiansen, 1985a; Guérard et al., tle aggressiveness can be considered to be highest for the true 2000; Horntvedt et al., 1983; Kim et al., 2008; Kirisits, 1998; parasitic species living permanently on healthy trees, interme- Krokene, 1996; Krokene and Solheim, 1998; Neal and Ross, diate for near obligate and facultative parasites, and very low 1999; Ross and Solheim, 1997; Solheim et al., 1993; Solheim to inexistent for the saprophytic species. For the aggregative and Långström, 1991; Yamaoka et al., 1995), intrinsically sup- species using the strategy of overcoming host resistance, the posing that a high beetle aggressiveness could be due to a high level of aggressiveness can be more precisely quantified by fungal pathogenicity. However, as results accumulated it soon the attack density above which trees are killed, i.e. the level became appearant that no correlation existed between fungal of the critical threshold of attack density (Berryman, 1976; pathogenicity and beetle aggressiveness (Harrington, 1993a,b, Christiansen et al., 1987; Lieutier, 2004; Paine et al., 1997; 2005; Lieutier, 2002; Paine et al., 1997). This has led several Raffa and Berryman, 1983a), aggressiveness being inversely authors to deny any role of Ophiostomatoid fungi in beetle related to threshold level. Such a threshold has been quantified establishment on conifers (Harrington, 1993b, 2005; Klepzig for some bark beetle species and, although it is influenced by and Six, 2004;Six,2003). genetic and environmental factors affecting the trees’ level of The reasons for controversy seem to result from misin- resistance, it can be used to rank these species according to terpretations. Indeed, studies using fungal pathogenicity to their aggressiveness. For example, the critical threshold of at- 2 demonstrate the role of Ophiostomatoid fungi in beetle estab- tack density, expressed in number of attacks per m of bark, lishment have been based on three kinds of arguments: was estimated, depending on tree vigor, between 50 and 120 for D. ponderosae on lodgepole pine (Raffa and Berryman, 1983a;Raffa, 2001; Waring and Pitman, 1983), around 75 for – assertion: tree death is a prerequisite for beetle establish- D. rufipennis on Picea engelmanii (Raffa, 2001), around 45 ment; for S. ventralis on Abies grandis (Raffa and Berryman, 1982a) – observations: most bark beetle species vector staining and between 200 and 400 for I. typographus on Norway spruce fungi and the sapwood of bark beetle killed trees is stained; (Mulock and Christiansen, 1986), while on Scots pine, that – experiments: experimental demonstrations that fungi iso- of I. acuminatus was around 850 (Guérard et al., 2000)and lated from aggressive beetles are able to kill trees after ar- that of Tomicus piniperda during its trunk attacks around 400 tificial mass inoculations supposed to mimic bark beetle (Långström and Hellqvist, 1993; Långström et al., 1992). attacks. Fungal pathogenicity is the ability to kill a tree, and the level of pathogenicity corresponds to the fungal virulence. In However, all these arguments do not lead to the conclusion that the case of bark beetle associated fungi, it has often been mea- fungi are responsible for killing the bark beetle attacked trees sured by the critical threshold of inoculation density, which before beetle attacks succeed. They do not even prove that is the density of artificial inoculations above which trees are fungi are involved in the tree killing process at all (Lieutier, killed (Christiansen et al., 1987; Lieutier, 2004; Paine et al., 2002). Indeed it is possible that fungi invade the sapwood af- 1997;Raffa and Berryman, 1983a). A low threshold indicates ter the tree has been killed by other factors. Moreover, artificial a high level of pathogenicity. In such artificial mass inocula- mass inoculations certainly mimic natural bark beetle attack tions, fungal cultures have generally been introduced at the very badly, as the conditions of fungal introductions differ cambium level, in holes previously bored with a puncher. As largely between the two situations (see below). In addition, it tree death is generally thought to result from disruption of has already been reported that trees can be successfully col- water transport in the sapwood, tree death caused by fungal onized and killed by bark beetles without blue stain fungi mass inoculations has been estimated by the fading of the fo- (Bridges et al., 1985; Whitney and Cobb, 1972). Nevertheless, liage or, most often, by measuring sapwood occlusion or sap- that fungi would not be responsible for killing trees attacked wood invasion by the fungus (percentage of occluded or blue by bark beetles would not prove either that they do not play stained sapwood in transversal bole sections), after harvest- any role in beetle establishment. One may even question the ing the trees a few months after inoculations (Christiansen, necessity that the trees be killed for beetle establishment to 1985a; 1985b; Christiansen et al., 1987; Croisé et al., 1998a; begin. Solheim et al., 1993, among others). In some cases, tree death

801p5 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al. has been determined by measuring sapwood specific hydraulic for fungal pathogenicity are indicated in Table I.Clearly,even conductivity of bole sections (Guérard et al., 2000). Some- when considering only Ophiostomatoid fungi associated with times, the percentage of dead phloem or cambium was also conifer bark beetle, there is no relation between beetle aggres- used (Krokene and Solheim, 1998). The same methods have siveness and fungus pathogenicity, a conclusion already pro- also often been used to compare the resistance level of dif- posed by Harrington (1993a; 1993b) and Paine et al. (1997). ferent trees to the same fungus or the effect of a treatment on There are three main possible explanations to such an obser- tree resistance (Brignolas et al., 1998; Christiansen and Glosli, vation: 1996; Sallé et al., 2008; Solheim et al., 1993, among oth- 1. Fungi do not play any role in beetle establishment and the ers) since the critical threshold of inoculation density depends ff association is not a real one, except in some cases when on genetic and environmental factors a ecting the trees’ level fungi provide complementary nutrition to the larvae; of resistance. Critical thresholds of inoculation density have 2. Fungal pathogenicity is not the right parameter to appre- been determined for various fungi. Expressed in the number 2 ciate the benefit brought by the fungus to beetle establish- of inoculations per m of bark, it is for example 100 to 200 ment on conifers (independently of possible food advan- for Ceratocystis polonica on Norway spruce (Christiansen, tage); 1985b). It is 300 to 400 for Leptographium wingfieldii (Croisé 3. Early tree death due to fungi is not a prerequisite for beetle et al., 1998a; Langström et al., 1993; Solheim et al., 1993), establishment. 400 to 800 for Ophistoma minus (Langström et al., 1993; Solheim et al., 1993), and more than 1000 for Ophiostoma We think that each of these explanations could be valid, de- brunneo-ciliatum (Guérard et al., 2000), all on Scots pine. In pending on the attack strategy used by the beetle species. Both many cases however, fungal pathogenicity has been tested on explanations 1 and 3 are valid for beetles using the strategy of seedlings with few inoculations, which is difficult to compare avoiding tree defenses. In such a strategy, the tree is not killed with mass inoculations on adult trees. and, in order to minimize the development of the hypersensitive reaction (see Sect. 2.1), the association with a fungus able to stimulate it, must be avoided (Lieutier, 1992; 2002). When 3.2. Relationships between beetle aggressiveness and beetles use the strategy of overcoming tree resistance, we think fungal pathogenicity that both explanations 2 and 3 are correct and that fungi effectively do play a role in beetle establishment. In the following Fungi associated with conifer bark beetles appear to have part of this paper, we will concentrate on this latter strategy diversified levels of pathogenicity but all of them, even those and try to explain why explanations 2 and 3 are valid. We will associated with aggressive tree killing bark beetles, need then report on results related to the modalities of fungal impact high inoculation densities to kill a tree. In contrast, classical on tree defenses, a point of importance to appreciate the real Ophiostomatoid pests not specifically associated with conifer role of fungi. bark beetles, such as Ophiostoma ulmi and Ophiostoma novo- ulmi in elms, Ceratocystis fagacearum in oaks, Ceratocystis 3.3. Fungal pathogenicity is not the right parameter to fimbriata in various woody plants, Leptographium wageneri appreciate the benefit brought by fungi to beetle in conifers, can kill a tree with only a few inoculations. With establishment on trees this in mind, it may be of interest to mention that most of these fungi are invasive species, thus corresponding to recent The crucial aspect in the beetle strategy to overcome host tree-fungus associations. Such a comparison leads to the state- resistance is to exhaust tree defenses or at least to considerably ment that conifer bark beetles are associated with moderately lower them so that the tree cannot put up effective resistance pathogenic fungi that have a very little capacity to colonize liv- to the aggressors and cannot impede beetle establishment and ing trees, as already concluded by several authors (Harrington, larval development. In this strategy, everything stimulating the 1993a; 2005; Lieutier, 2002; Paine et al., 1997;Raffaand tree’s defense reactions accelerates the tree’s energy expen- Klepzig, 1992, among others). diture and thus lowers the critical threshold of attack den- In these conditions, one may wonder if a relationship be- sity (Lieutier, 1992; 2002). Introducing fungi able to strongly tween beetle aggressiveness and fungus pathogenicity exists stimulate the tree defensive reactions, into the beetle galleries, among blue stain fungi associated with conifer bark bee- would thus be of great help for beetle populations, since it tles. Table I presents the relationships between conifer bee- would allow them to establish on trees at a lower population tle species and Ophiostomatoid fungal species supposed, for level than without fungi (Berryman, 1972; 1976; Christiansen most of them, to play a role in beetle establishment. Beetle et al., 1987; Franceschi et al., 2005; Lieutier, 2002; Paine et al., species are grouped according to their level of aggressiveness 1997;Raffa and Berryman, 1983a). Consequently for estab- and fungi to their level of pathogenicity. Because the critical lishment, the beetle population doesn’t need fungus to kill the threshold of attack density has been determined for very few tree but to help in overcoming tree resistance through rapidly species only, bark beetle aggressiveness has been estimated stimulating host defenses (by causing extensive reactions) in according to the importance of damage on living trees, es- each place where it can be inoculated (Lieutier, 2002;Paine pecially on healthy trees, mainly by using information from et al., 1997). In this context, there is no reason for an a priori Furniss and Carolin (1977) in North America, and Chararas relationship between fungus ability to stimulate tree defense (1962) and Grégoire and Evans (2004) in Europe. References reactions when introduced into the tree by a beetle, and its

801p6 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Table I. Relationships between various conifer beetle species and their Ophiostomatoid fungi, while grouping beetles according to their level of aggressiveness [Furniss and Carolin (1977) and Dymerski et al. (2001) for North America; Chararas (1962) and Grégoire and Evans (2004) for Europe] and fungi according to their level of pathogenicity. H = High, M = moderate, L = low aggressiveness/pathogenicity.

H M L (5) (37) (7, 12, 20) 12, (7,

( 10, 18) ( 10, (43)

(25, 30, 37) i (41) (14, 28) um (33, 34, 42) 34, (33, (36, 44) (8, 16, 27) 16, (8,

e m (30, 32, 37) t (10, 36, 44) (10, (35, 37, 39) 37, (35, um a

t a i (14, 16, 22) ioides a tis (3, 9, 11, 12, 42) 12, 11, (3, 9, um um (12, 13,34) 33, (14, 33, 34, 42) 34, 33, (14, ico h t

uga e (4, 17, 19, 20, 38, 39) ti er nn v a biotic an ava e anum a a num n g a i o ic rop ricicol icl o ym eo-cili dots n u fipe a m av (2, 6, 20, 23, 24) 13, n icill O. c O. u i u bieti ng n (1, 19, 20, 22, 26, 29, 31, 38, 40) 31, 38, 40) 29, 26, 20, 22, 19, (1, fieldii fieldii a O. bre a A. s O. . cl unn s 0. pice C. l O. e ng C. r G L. terebr L. L. i O. O. bicolor nu O.pe i O. lo O. O. ips w C. polo O. pse m O. br O. L. L. O.

I. typographus X X X X X X D. ponderosae X X X D. frontalis X D. brevicomis X X H D. pseudotsugae X X X X D. rufipennis X X D.micans X S. ventralis X

I. cembrae X X X I. sexdentatus X X X X I. acuminatus X X X X M P. chalcographus X X X O. erosus X I. pini X

T. minor X D. valens X X L D. terebrans X T. piniperda X X

References related to the level of fungal pathogenicity or the association with beetle species: 1 = Basham (1970); 2 = Ben Jamaa et al. (2007); 3 = Christiansen (1985a); 4 = Croisé et al. (1998a); 5 = Filip et al. (1989); 6 = Francke-Grosmann (1963); 7 = Guérard et al. (2000); 8 = Highley and Tattar (1985); 9 = Hornvedt et al. (1983); 10 = Kim et al. (2008); 11 = Kirisits (1998); 12 = Kirisits 1999); 13 = Kirisits (2004); 14 = Kirisits et al. (2000); 15 = Klepzig et al. (1995); 16 = Krokene and Solheim (1998); 17 = Långström et al. (1993); 18 = Lee et al. (2006); 19 = Lieutier et al. (1989b); 20 = Lieutier et al. (1990); 21 = Lieutier et al. (2005); 22 = Mathiesen (1950); 23 = Mathre (1964a); 24 = Mathre (1964b); 25 = Neal and Ross (1999); 26 = Owen et al. (1987); 27 = Rane and Tattar (1987); 28 = Redfern et al. (1987); 29 = Rennerfelt (1950); 30 = Ross and Solheim (1997); 31 = Siemaszko (1939); 32 = Six and Bentz (2003); 33 = Solheim (1992b); 34 = Solheim (1993b); 35= Solheim (1995); 36 = Solheim and Krokene (1998a); 37 = Solheim and Krokene (1998b); 38 = Solheim and Långström (1991); 39 = Solheim and Safranyik (1997); 40 = Solheim et al. (1993); 41 = Solheim et al. (2001); 42 = Viiri and Lieutier (2004); 43 = Whitney and Cobb (1972); 44 = Yamaoka et al. (1995).

801p7 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al. ability to kill a tree, especially when the latter is determined 1983a). The physiological status of the tree must also be taken through the critical threshold of artificial inoculation density. into account since it can condition the phloem response in var- In fact, mass inoculations allow a comparison of the ious ways (Croisé and Lieutier, 1993;Croiséetal.,1998b; pathogenicity levels of different fungi or the level of resis- Sallé et al., 2008; see also references in Paine et al., 1997; tance of different trees to be made, but they certainly mimic and Lieutier, 2004). Modalities of fungal impact on tree de- bark beetle attacks very badly. With artificial mass inocula- fenses should also be considered when interpreting the data tions, every wound is inoculated with the fungus, which is (see Sect. 3.6). often far from being the case during a beetle mass attack, as Direct comparisons between phloem reaction zone length the frequency of beetle contamination is rarely 100% and can after low density inoculations and fungus pathogenicity af- even be extremely low in certain populations: see for exam- ter mass inoculation (or sapwood invasion after beetle at- ple Kirisits (2001), Viiri and Lieutier (2004) and Sallé et al. tacks) have shown that pathogenicity is not related to fungus (2005)forI. typographus with C. polonica, Bridges et al. ablity to stimulate tree defenses (Krokene and Solheim, 1997; (1985)forD. frontalis with O. minus, Lieutier et al. (1989b) 1998; 1999; Solheim, 1988; 1992a; 1992b). As an example, and Solheim and Långström (1991)forT. piniperda with L. in Figure 1, Norway spruce trees were inoculated with four wingfieldii, Lee et al. (2006)forD. ponderosae with O. mon- fungal species, either at a low density (see left part of the fig- tium and G. clavigera. The number of spores introduced by a ure) or at a high density (see right part of the figure) (Krokene beetle in its gallery may also differ largely from that contained and Solheim, 1999). Among all assayed fungal species, only in an inoculated disc of sporulated culture (see below). Gallery C. polonica was really pathogenic (as indicated by its ability to excavation by beetles contributes to fungal spread inside the stain the sapwood after mass inoculations), whereas all species host phloem, which is not the case for artificial inoculations were able to stimulate the phloem reaction significantly after (Raffa and Klepzig, 1992). Moreover, it has been reported that low density inoculations (as indicated by the extension of the the pattern of host colonization by fungi after artificial mass phloem reaction zone). It thus clearly appeared that the pat- inoculations differs from that observed following beetle mass tern of fungal ability to stimulate the phloem reaction differed attack (Parmeter et al., 1992). largely from the pattern of fungal pathogenicity. Another sig- Since the important factor is the ability of the fungi to stim- nificant example is Ambrosiella symbioticum, a fungus associ- ulate tree defenses, results of inoculation experiments aiming ated with S. ventralis. It strongly stimulates A. grandis phloem at appreciating the role of the fungus in beetle establishment reactions after low density inoculations (Raffa and Berryman, must be expressed in terms of tree defense stimulation rather 1982a; Wong and Berryman, 1977) but is unable to penetrate than in terms of killing trees. In such a way, mass inocula- the sapwood after mass inoculations (Filip et al., 1989). It tions can be useful to assess the stimulation ability of various should be noted however that, when comparing isolates from fungi or to examine the mechanisms of defense exhaustion, the same species, a correlation between reaction zone length when approaching the critical threshold as done for example in after low inoculation density and fungal pathogenicity can be the case of G. clavigera on P. contorta (Raffa and Berryman, observed (Lieutier et al., 2004; Plattner et al., 2008; Sallé et al., 1983b). In addition to mass inoculations, low density inocula- 2005). tions (usually less than 20 inoculations per tree) far below the Moreover, the fungus invading the sapwood after a beetle critical threshold of inoculation density, are also very useful attack is not necessarily the one that plays a major role in stim- to evaluate fungus ability to stimulate tree defenses, especially ulating tree defenses, even after high inoculations densities. In because the tree stays alive and can better exhibit the intensity Figure 2,fivedifferent blue stain species are compared regard- of its response to aggression. Such inoculations allow under- ing their ability to stimulate the phloem hypersensitive reac- standing basic mechanisms of tree defenses and the processes tion of Norway spruce after high density inoculation (Solheim, leading to their stimulation, prior the alterations due to high 1988), and their ability to grow into the sapwood after a suc- inoculation densities (see for example Brignolas et al., 1995; cessful natural infestation by I. typographus (Solheim, 1993a). Franceschi et al., 2005; Kim et al., 2008; Lieutier et al., 1995, Whereas O. penicillatum induced more extensive phloem re- 1996). Moreover, different fungal species and strains can be action zone than C. polonica and O. bicolor, it stayed largely compared on the same trees, thus limiting variations due to behind these latter two species during sapwood invasion. The trees (Rice et al., 2007a). It has also been suggested that tak- pattern of the two fungal behaviors thus completely differed. ing into account the speed of reaction development is of im- The pattern of fungal growth also largely differs between portance for the interpretation of the results from low density the phloem and the sapwood. In Figure 3 for example, four inoculations (Wallin and Raffa, 2001). The two approaches blue stain species were compared for their ability to grow thus complement each other. However, interpretations of re- in the tissues of Douglas fir after low density inoculations sults from artificial inoculations must be done carefully. In all (Solheim and Krokene, 1998b). Ophiostoma pseudotsugae cases, adult trees, i.e. comparable to trees attacked by the bee- and O. europhioides extended much more than L. abietinum tles, must be used. Fungus stimulation capability must also and C. rufipenni in the phloem, whereas no difference between be estimated after a delay comparable to that involved for the species was observed in the sapwood although the two lat- issue (success or failure) of beetle attacks to be determined, ter species tended to grow deeper. Clearly and as already re- which is often within a few days (from 4–5 to 20) depending ported, various blue stain species mainly infest the phloem and on beetle aggressiveness and weather conditions (Anderbrandt show a slow spread in the sapwood, whereas the opposite is et al., 1988; Lieutier, 2002; Payne, 1980;Raffa and Berryman, true for other species (Solheim, 1992b; Solheim and Krokene,

801p8 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Low density inoculations High density inoculations

Phloem reaction zone length (mm) after 5 weeks Bluestained sapwood (%) after 15 weeks 40 60 ab a bc 30 c 40

20 d

20 10

0 0

1 2 3 4 5

C C

D D

Figure 1. Direct comparisons between phloem reaction development after low density inoculations (three inoculations per fungal species in each tree) and fungus pathogenicity after mass inoculations (400 inoculations per m2 of a given fungus per tree), in Norway spruce (Picea abies) inoculated with various blue stain fungal species (adapted from Krokene and Solheim, 1999). Values with the same letter did not diﬀer signiﬁcantly.

0 Growth delay in the sapwood: Distance (mm) behind the leading edge of fungal invasion

Successful natural infes- 10 Phloem lesion length 60 taon by I. typographus aer 10 weeks (mm) 50 a b 20 Arﬁcial high density 40 inoculaons c 30

20 d

10 no data 0 1 2 3 4 5

Figure 2. Extension of the phloem hypersensitive reaction zone after high density (475/m2) inoculations of Norway spruce (Picea abies) with various fungal species, compared with fungal growth of the same species in the sapwood of Norway spruce after natural infestations by I. typographus (adapted from Solheim, 1988, 1993a). Values with the same letter did not diﬀer signiﬁcantly.

1998a,b; Uzunovic and Webber, 1998). Such different behav- wood however, C. rufipenni induced a more extended reac- iors may be explained by referring to oxygen availability in tion zone than O. pseudotsugae and O. europhioides, while tree tissues, the ability to invade sapwood being possibly re- the reaction zone induced by L. abietinum was more extended lated to high tolerance to oxygen deficiency (Solheim, 1991; than that of O. pseudotsugae. One can thus think that during 1992a; Solheim and Krokene, 1998a). a beetle attack, both fungal categories certainly interfere in Figure 3 also shows that for all species, stimulation of the stimulation of tree defenses, those growing better in the tree defenses concerned both the phloem and the superfi- phloem stimulating mainly phloem defenses, and those grow- cial sapwood. Between fungi comparisons related to reaction ing better in the sapwood stimulating mainly superficial sap- zone extension in the phloem show a tendency similar to that wood defenses. Meanwhile, as it appears in Figure 3 related of fungus extension in the phloem. In the superficial sap- to Douglas fir, stimulation of tree defenses seems mainly a

801p9 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

Phloem Sapwood 100 % blue stained sapwood

C O. pseudotsugae 80 b a

B 60 a a O. europhioides

AB L. abietinum 40 c a

A C. rufipenni d a 20

-60 -50 -40 -30 -20 -10 0 01020 0 Extension of the tree defense reaction zones after 3 weeks 10 100 1000 10000 Fungal growth after 3 weeks Number of beetle attacks

Figure 3. Fungal growth and extension of the tree defense reaction Figure 4. Development of the blue stained area in the sapwood of zones in the phloem and in the sapwood of Douglas fir (Pseudotsuga Norway spruce (Picea abies), with increasing numbers of I. typogra- menziesii), three weeks after artificial low density inoculations with phus natural attacks (adapted from Christiansen, 1985b). various fungal species (three inoculations per fungal species in each tree). Results of statistical tests were not available for phloem reaction zone extension (adapted from Solheim and Krokene, 1998b). In the come, i.e. before tree is killed. Paine et al. (1997)haveal- same tissue, values with the same letter did not differ significantly. ready concluded that trees are overcome well in advance of fungal growth in sapwood or changes in tree moisture status. By comparing the timing of increase in stem circumference phloem phenomenon, especially when considering only the with that of variations in sap flow velocity of loblolly pines fungal species associated with the Douglas fir beetle (and thus naturally colonized by D. frontalis, Wullschleger et al. (2004) excluding C. rufipenni). Evaluation of energy expenditure in also reached the conclusion that disruption of water balance building tree defenses is needed in each tissue to really com- was not a prerequisite for beetle oviposition and larval devel- pare the role of fungal species in exhaustion of tree defenses. opment. In these conditions, sapwood (and phloem) invasion In addition, a possible interference of other micro-organisms would simply be the unavoidable consequence of the consid- such as bacteria should also be taken into account (Cardoza erable depletion of tree resistance. Considering that beetles are et al., 2006). However, it is worth mentioning that stimula- often associated with a complex group of fungi rather than tion of sapwood defenses concerns only the superficial part with a single fungal species (see below, Sect. 4.1), at the mo- of this tissue. This is an additional reason for not using fun- ment when tree resistance is overcome, any present fungus, gal pathogenicity and sapwood invasion to make conclusions even with a low level of pathogenicity, depending on substrate about the role of fungi in beetle establishment. characteristics and on its competitive ability, could invade the sapwood successfully and thus possibly be involved in tree death. Similarly, any insect present could also succeed in es- 3.4. Prior tree death due to fungi is not a prerequisite tablishing in the phloem. Death would then result from all in- for beetle establishment vaders acting together (Berryman, 1972; Lieutier, 2002;Paine et al., 1997), in addition to the sacrifice of tissues through the This statement is, in some ways, the consequence of the induced reactions (Lieutier, 2002), and would occur after the lack of relationship between fungus ability to stimulate tree completion of the critical interactions between the tree and its reaction and fungus pathogenicity. As long as tree resistance aggressors (Franceschi et al., 2005; Lieutier, 2002; Paine et al., is not overcome (below the critical threshold of attack density), 1997). it contains the aggressors. Fungi can extend into the sapwood The fastest growing species in the sapwood may have a only very little or not at all while beetles cannot establish in dominant role in tree death, in agreement with the sugges- the phloem. As soon as tree defenses are exhausted (above the tion by Paine et al. (1997) that competition with other fungi threshold), nothing impedes the development of the aggres- could have been a driving force of pathogenicity. However, sors and, even if a low level of resistance may still persist, the presence of a fungus in the sapwood does not prove that phloem and sapwood are invaded rapidly. As an example, Fig- it is responsible for tree death since sapwood at this moment ure 4 shows that, during a natural attack by I. typographus on can be colonized by any present pathogen. Several studies Norway spruce, no blue stain was observed in the sapwood have concluded that fungi are not the primary factor of mor- until a certain number of beetle attacks was reached but this tality in trees attacked by bark beetles (Nebeker et al., 1993; tissue was invaded rapidly above that threshold (Christiansen, Parmeter et al., 1992;Stephenetal.,1993). Hobson et al. 1985b). If, as generally thought, tree death effectively results (1994) have even reported that during attacks by D. brevi- from disruption of water transport in the sapwood, the trees comis on Pinus ponderosa, Ophiostoma species apparently in- were still alive when the threshold was reached. We thus think vade the sapwood after the tree has died. In addition, there that beetle establishment begins after tree resistance is over- are several examples where trees were killed without fungus.

801p10 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Galleries with Ophiostoma phloem) and fungi (in the sapwood and phloem). Experiments Ophiostoma with beetle attacks at low density have demonstrated that fungi Galleries without are eﬀectively able to stimulate tree reaction when introduced by a beetle into the tree. Figure 5 gives an example for I. sex- Reaction zone length (mm) 70 dentatus on Scots pine, after beetle attraction to the trees with a' synthetic pheromones (Lieutier et al., 1995). Phloem reaction 60 a zones were signiﬁcantly more extended in front of galleries with Ophiostoma than in front of those without Ophiostoma.

50 The observation that D. frontalis attacks without fungi succeed at higher densities than attacks with fungi (Bridges et al., 1985; Whitney and Cobb, 1972) is also in agreement with the 40 b role of fungi in helping overcoming tree resistance. b'

30 a" 3.6. Modalities of fungal impact on tree defenses

20 Several parameters can affect the role of the fungi during a b" beetle mass attack, which must be taken into account before 10 concluding about the role that is or isn’t played by a fungus in beetle establishment. These factors have already been mentioned as invalidating the use of artificial mass inoculations 0 60 12 87 10 8 6 followed by measuring fungus pathogenicity to appreciate the 123 Tree 39 Tree 40 Tree 98 benefit brought by the fungus to the beetle. They also condition the utilization of the results from artificial inoculations (at Figure 5. Phloem reaction zone length in three Scots pine (Pinus low or high densities) performed to appreciate beetle ability to sylvestris) trees in front of beetle galleries, 6 to 9 weeks after I. sex- stimulate tree defenses. They are the frequency of beetle con- dentatus attacks associated or not with an Ophiostoma species. The tamination, the diversity of fungal species, the between-isolate number of galleries is indicated under the histograms (adapted from variability in a fungal species, and the level of fungus load of Lieutier et al., 1995). Values with the same letter did not differ signif- a contaminated beetle. icantly. – The frequency of beetle contamination by a fungus can vary largely between localities (even at a relatively short distance) as well as with time in a same locality, and such huge That was reported in the case of D. frontalis (Bridges, 1983; variations certainly affect the role of the fungus in beetle es- Bridges et al., 1985; Hetrick, 1949). Similarly, pines have been tablishment. killed by D. ponderosae, although G. clavigera, the only D. In Europe, C. polonica has been reported as a common ponderosae-associated fungus able to kill trees after mass in- associate of I. typographus in several localities (Harding, oculation, was absent (Six and Klepzig, 2004). 1989; Kirisits, 2001;Siemaszko,1939; Solheim, 1986). It was however found less frequently and even very rarely in other places (Harding, 1989; Jankowiak, 2005; Kirisits, 2001; 3.5. Fungi do play a role in beetle establishment on trees Mathiesen-Käärik, 1953; Sallé et al., 2005; Solheim, 1993b; Viiri, 1997; Viiri and Lieutier, 2004). Similar situations have We think that the beetle strategy of being associated with been observed in North America with the associations between moderately pathogenic fungi is really in operation in case D. rufipennis and C. rufipenni (Six and Bentz, 2003;Sixand of the strategy of exhausting tree defenses, and that it acts Klepzig, 2004)orL. abietinum (Aukema et al., 2005), between through the ability of the fungi to lower the critical thresh- D. ponderosae and G. clavigera or O. montium (Six and Bentz, old of beetle attack density by stimulating tree reactions to ag- 2007; Six and Klepzig, 2004), and between D. frontalis and gressions (Franceschi et al., 2005; Lieutier, 2002; Paine et al., O. minus (Bridges et al., 1985). The consequences, for bee- 1997). This association is even a necessity for successful beetle population dynamics, of huge variations in the frequency tle attacks in trees with high resistance level. It has been sug- of fungal contamination have already been evoked through the gested that nutrition is very likely an important benefit for bark effects on beetle nutrition and reproduction (Six and Bentz, beetles (Adams and Six, 2007; Bleiker and Six, 2007; Klepzig 2007; Six and Klepzig, 2004). Certainly such variations in- and Six, 2004;Six,2003). We believe that stimulation of the terfere in beetle population dynamics also through effects on tree defense reactions is also a real benefit brought by the helping beetle establishment on trees. fungi to the beetles. Following Paine et al. (1997) and Lieutier – Additionally, the diversity of fungal species associated (2002), beetle establishment is suggested to proceed in two with a beetle population must also be considered. There are successive steps: (1) overcoming tree resistance (exhausting many examples of multiple associations in bark beetle species tree defenses by stimulation) in both the phloem and the su- (Lee et al., 2006;Six,2003; Six and Paine, 1999;asexam- perficial sapwood; (2) Invasion of tree tissues by beetles (in the ples for North America, Kirisits; 2004, and references therein

801p11 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

180 a Phloem reaction zone length (mm) ab ab ab ab 160 ab ab Fungal growth in the phloem (mm) ab ab abc 140

abc 120 abc bc bc

100

A c A A 80 A A A A AB AB AB 60 BC BC BC CD D D D 40 D

0 225801 683 1214 1267 603 304 41291 63 439 384 1123554 1151 Isolates

Figure 6. Phloem reaction zone length and fungal growth, 14 days after low density inoculations in Scots pine with 15 different isolates of L. wingfieldii from the Orléans forest (two inoculations per fungal isolate in each tree) (adapted from Lieutier et al., 2004). Values with a same letter did not differ significantly. for Europe). In forests of southern Poland, the frequency of the case of other beetle species. Variations of associations be- I. typographus beetles carrying O. polonicum with was 5.6%, tween D. frontalis and its symbiotic fungi, as well as between while it was 54.8% for O. penicillatum, 30.8% for O. bicolor, beetle-associated-mites and fungi, have also been reported to 27% for O. ainoae, 23.3% for O. piceaperdum and 20.3% occur in relation to temperature (Hofstetter et al., 2007). for O. piceae (Jankowiak, 2005). In these conditions, even if – The ability to stimulate tree defenses can also vary among O. polonicum appears the most pathogenic and even if it is isolates within the same fungal species. A comparison be- able to strongly stimulate tree responses after artificial inoc- tween 15 isolates of L. wingfieldii collected on the same date ulations, one may doubt its dominant role in beetle establish- within the same forest revealed large variations in their abil- ment (and even in sapwood invasion) in the natural conditions ity to stimulate phloem tree defenses and to grow into the of these forests. Raffa and Smalley (1988) mentioned that “the phloem of Scots pine, after artificial low density inoculations relative abundance and proportions of various components of (Fig. 6) (Lieutier et al., 2004). The locality had no effect and, the fungal flora could have a strong effect on the colonization owing to the dispersal capabilities of the vector T. piniperda, success”. To appreciate the role of fungi in the establishment it was concluded that all isolates were very likely to coexist of a beetle population, it may thus be more important to con- throughout the forest. They thus certainly coexist also within sider the total percentage of beetles carrying blue stain fungi the same beetle population. Similar results have been obtained rather than the percentage of beetles associated with the fun- among isolates from different localities, with O. bicolor and gus which is most efficient in stimulating tree defenses. The O. piceaperdum associated with I. typographus (Sallé et al., frequencies of the associations of D. ponderosae with G. clav- 2005) and with G. clavigera associated with D. ponderosae igera and O. montium can vary largely, one or the other fungal (Plattner et al., 2008;Riceetal.,2007a). In the latter associa- species being the most prevalent depending on localities and tion, variations among isolates in their adaptation to cold tem- time (Adams and Six, 2007; Bleiker and Six, 2009;Sixand perature have also been reported (Rice et al., 2008). Such phe- Bentz, 2007) probably in relation to differential performances nomena could have evident consequences regarding the role of of each species under different temperature or moisture condi- the associated fungus in stimulating tree defenses and conse- tions (Bleiker and Six, 2009;Riceetal.,2008; Six and Bentz, quently in beetle attack success. 2007). However, as the two fungal species play a similar role – The level of fungus load carried by an individual bee- in beetle nutrition (Adams and Six, 2007), they are believed tle must also be considered to appreciate the real role of an to complement each other, and the prevalence of one or the associated fungus in stimulating tree defenses in the event of other is said to assure the insect a correct feeding substrate natural beetle attacks. Experiments with artificial low density any time in all habitats of its geographic range (Bleiker and inoculations have demonstrated that resin concentration in the Six, 2009;Riceetal.,2008; Six and Bentz, 2007). Comparable phloem hypersensitive reaction zone that develops around a complementarities could be suggested between fungal species wound is positively related to the number of spores introduced, involved in the stimulation of tree defenses, in this case and in and that a minimum number of spores must be present to cause

801p12 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

mg resin / g fresh phloem in the first 25 mm of the phloem reaction zone, 5 weeks after inoculation 350 350

300 L. wingfieldii 300 O. brunneo-ciliatum

250 250

200 200

150 150

100 100

50 50

0 0 12345678 12 34 56 78 9

Numberof spores in sterile water

not significantly different from significantly different from inoculation of a 5 mm diameter sterile water alone sterile water alone diskof malt agar culture

Figure 7. Resin concentration in the phloem hypersensitive reaction zone of Scots pine, 5 weeks after low density inoculations (16 to 18 inoculations per tree in total) with sterile water containing various numbers of fungal spores of L. wingfieldii and O. brunneo-ciliatum,and comparison with inoculations with 5 mm diameter disks of 3-week-old malt agar cultures (adapted from Lieutier et al., 1989a). Statistical tests were performed by comparing each resin concentration value to that obtained after inoculation with sterile water alone, and their results are indicated by the histogram color. For a tree response to be significantly different from that induced by sterile water, at least 104 spores were needed in the case of L. wingfeldii, and 102 in the case of O. brunneo-ciliatum. a significant response (Lieutier et al., 1988; 1989a). For exam- 4. WHY AND HOW BEETLES-FUNGI ple (Fig. 7), this minimum number was higher than 104 spores RELATIONSHIPS INTERFERE IN THE for L. wingfieldii, a fungus vectored by T. piniperda, whereas STRATEGY OF EXHAUSTING TREE DEFENSES it was 100 spores only for O. brunneo-ciliatum, vectored by Here we discuss beetle – fungi relationships on a wider tem- I. sexdentatus. Certainly a contaminated I. sexdentatus beetle poral scale, by proposing functional and evolutionary hypothe- can introduce 100 spores of O. brunneo-ciliatum in its gallery ses for the role of Ophiotomatoid fungi in beetle establishment when attacking a tree, but it is not certain that a contami- through stimulating tree defenses. We propose: (1) that any nated T. piniperda beetle carries a sufficiently high number 4 beetle species can develop an association with a fungus species of L. wingfieldii spores to introduce at least 10 of them in its belonging to the Ophiostomatoid flora of its host tree, and can gallery. A 5 mm diameter disc of a 3-week-old sporulated agar take advantage of this association for its establishment; (2) that culture of L. wingfieldii or O. brunneo-ciliatum (that is com- the selection pressure that has led fungi to develop their abil- parable to the discs commonly used in most experiments us- 6 ity to stimulate tree defenses is the necessity of a considerably ing artificial inoculations) contains at least 10 spores (Lieutier low level of tree resistance for fungus extension into the tree. et al., 1989a) and induces a tree response comparable to that Meanwhile, we replace the stimulating role of the Ophiostom- obtained with such a high number of spores (Fig. 7). This atoid fungi in the general context of conifer-bark beetle-fungi could explain why L. wingfieldii, although strongly stimulat- relationships, leading us to present our views on the general ing tree reactions and having a high pathogenicity with artifi- functioning of these relationships and finally to hypothesize cial inoculations (Croisé et al., 1998a; Lieutier et al., 1989b; that beetle species are associated with fungal complexes, of Solheim et al., 1993; Solheim and Långström, 1991), is not which species assume various roles regarding their relation- able to stimulate the phloem hypersensitive reaction during ships with beetles and trees. T. piniperda trunk attacks (Lieutier et al., 1989b; 1995)and thus not able to help this beetle in its establishment on the trunk. On the contrary, O. brunneo-ciliatum can play a real 4.1. Origin of the associations and existence of fungal role during I. sexdentatus attacks on the same tree species complexes (Lieutier, 1995; Lieutier et al., 1989a; 1989b; 1995). Using Beetle-associated fungal flora is generally similar between artificial inoculations of fungal culture to appreciate fungus two conifer species belonging to the same genus, even at- ability to stimulate tree defenses may thus lead in some cases tacked by beetle species from different genera. Conversely, to overestimate the role of the fungus during natural beetle there are often large differences between the beetle-associated attacks. fungal floras of conifers belonging to different genera, even

801p13 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al. attacked by closely related beetle species. As examples in Eu- no coevolution has occurred between beetles and Ophiostom- rope, the fungal flora associated with I. typographus, I. dupli- atoid fungi (Harrington, 1993a). Being basically adapted to catus and I. amitinus on spruce resembles that of P. chalcogra- trees, blue stain fungi have then adapted to transportation by phus (also attacking spruce) much more than that associated beetles that live on these trees because they benefit from being with I. sexdentatus and I. acuminatus on pines, which on the easily disseminated and then introduced, through wounding, contrary resembles that of O. proximus (also attacking pines) into a suitable host (Harrington, 1993a; 2005; Kirisits, 2004; (Kirisits, 2004). This suggests that the host tree has more im- Six, 2003). However, as said above, fungus extension in sap- portance than the beetle in speciation of Ophiostomatoid fungi wood is allowed only if tree resistance is depleted or at least (Harrington, 1993b; Kirisits, 2004). Phylogenetic studies of considerably lowered. the Ceratocystis species that develop on conifers have also We thus hypothesize that the necessity of a considerably concluded that adaptation of these fungi to trees is older than low level of tree resistance for fungus extension into the tree is their adaptations to bark beetles (Harrington and Wingfield, the selection pressure that has led fungi to develop their intrin- 1998). In these conditions, it is logical to suppose that fun- sic ability to stimulate tree defenses. In healthy vigorous trees gal ability to stimulate tree defenses is also adapted to the attacked by very aggressive beetles, fungi must have devel- host tree. oped a high ability to stimulate tree defenses. This selection We thus first hypothesize that any beetle species would be could have been induced, directly or indirectly, through the helped in its establishment on a given tree species by develop- ability of the fungus to grow into the phloem. Indeed, in a con- ing an association, even loosely, with a fungus species belong- text of interspecific competition, growth speed is a crucial ele- ing to the Ophiostomatoid flora of that tree species. In addi- ment for a fungus, since in both dead and living trees it condition, because several fungi often cohabit in a tree, we suppose tions fungus access to resources like food or ability to produce that the association could be developed with various species in fruiting bodies in the galleries of dispersing beetles, and fungal the same beetle population, each fungal species playing a com- extension is positively related to phloem reaction zone length parable or complementary role, even they aren’t as efficient. after low density inoculations (Ben Jamaa et al., 2007;Lieu- Such a comparable and complementary role has already been tier et al., 1989b; 2004; 2005; Wong and Berryman, 1977). suggested for G. clavigera and O. montium in their feeding as- That aggressive beetle species develop more frequently in liv- sociation with D. ponderosae (see Sect. 3.6). The beetle pop- ing trees than moderately or low aggressive species would thus ulation would thus be associated with a fungal complex. For have lead, for the former, to more specific associations with I. typographus as an example, a similar background of fungal fungi able to grow rapidly in living trees, and thus to strongly species is generally found, C. polonica, O. anoiae, O. bicolor, stimulate tree reactions. Moreover, in the context of an asso- O. penicillatum and O. piceaperdum being the most common ciation with bark beetles for transportation, it must be noted and ecologically important species (Kirisits, 2004), but vary- that developing the ability to stimulate tree defenses to in- ing in proportion depending on locality and time (see above vade a tree is a better strategy for a fungus than developing its Sect. 3.6). In addition, competition certainly occurs between own pathogenicity, because this latter choice would have led the different fungal species of a complex in a bark beetle habi- to competition with the insect for the tree tissues (see below). tat (Bleiker and Six, 2009; Klepzig et al., 2001a; 2001b). The efficiency of the beetle fungus associations in beetle popula- 4.3. The necessity of controlling fungus extension tion establishment would then depend on several components acting complementarily: the intrinsic ability of the beetle- Being associated at a high frequency with a fungus that ex- associated fungi to stimulate tree defenses, the percentage of tends quickly in the tree is a risk for beetle progeny if fun- contaminated beetles in the attacking population, the level of gus extension is not controlled after tree defenses are over- individual inoculum, and the possible competitive interactions come (Harrington, 1993a; Lieutier, 2002; Six and Klepzig, between fungal species. As each of these components can be 2004, among others). There are several examples indeed where affected by environmental factors and substrate variability, it is phloem invaded by fungi is unsuitable for beetle larvae, such not surprising that the nature of the dominant beetle-associated as in the case of O. minus for D. frontalis (Barras, 1970; fungal species and the efficiency of the insect-fungus asso- Klepzig et al., 2001a; 2001b), O. montium for D. ponderosae ciations vary considerably in space and time, especially for (Six and Paine, 1998)orO. ips for I. avulsus (Yearian, 1966; beetle species with large geographic distribution and high dis- in Klepzig and Six, 2004). These observations raised the ques- persal capabilities. Moreover, diversified environmental con- tion of how a fungus that is antagonist to the beetle larval de- ditions favor a considerable mixing between beetle individuals velopment can be maintained in a beetle population at a high and associated fungal flora, certainly resulting in limiting se- level of association, instead of being strongly counter-selected. lective pressures, which in return would favor the coexistence Klepzig et al. (2001a) and Six and Klepzig (2004) proposed of different fungal species in a same beetle population. that an explanation could be the benefits brought by such fungi to the phoretic mites carried by the beetles. Certainly also, the benefit brought by the fungus in stimulating tree defense reac- 4.2. Why fungi stimulate tree defenses tions is a valid reason. For the beetle, equilibrium must be maintained between According to the above hypothesis, the beetle-fungus asso- benefit from stimulating tree defenses and antagonism to lar- ciations can be weak and variable, which proves that certainly val development. Dendroctonus frontalis has achieved such an

801p14 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801 equilibrium by developing an association with a mycangial and beetles thus certainly exist. Their complexity is even in- species of the genus Entomocorticium, antagonist to O. mi- creased, at least in some cases, by additional interactions with nus in addition to bringing nutrients to beetle larvae (Klepzig other partners such as nematodes, bacteria and mites (Cardoza et al., 2001a; 2001b). Similar associations, eventually with et al., 2006; 2008; Klepzig et al., 2001a; 2001b). By interfer- other micro-organisms such as bacteria (Cardoza et al., 2006), ing in beetle establishment and brood development in trees, certainly also exist in other bark beetle species. In the fungal these complex interactions must play an important role in the complexes associated with bark beetles, there is often one rel- beetle population dynamics and outbreaks. atively virulent blue stain species, while others are less virulent (Kirisits et al., 2000; Kirisits, 2004; Lee et al., 2006; Lieutier et al., 1989b; Six and Klepzig, 2004; Solheim, 1988). 5. TYPES OF BEETLE-FUNGUS ASSOCIATIONS The role of each of them is ignored. Possibly antagonisms comparable to those described in the D. frontalis complex There is a large diversity of types of bark beetle – Ophios- also exist, resulting in beetle brood protection and nutrition. tomatoid fungus associations, from absent or very simple to Phloeomycetophagous species with sac mycangia should be very elaborate. In the following section, we tentatively classify concerned (Klepzig and Six, 2004;Six,2003; Six and Klepzig, these associations by using the above considerations. Among 2004), amongst which are D. ponderosae (Six and Paine, beetle species able to develop on living trees, we are able to 1998; Whitney and Farris, 1970), D. jeffreyi (Six and Paine, define three main groups while indicating the possible levels 1997), D. brevicomis (Paine and Birch, 1983)orI. acumi- of beetle aggressiveness within them. For species using the natus (Francke-Grosmann, 1967). Beetles with pit mycangia strategy of overcoming tree resistance, we also tentatively dis- may also be concerned especially if pits have glands, such as tinguish fungal species mainly stimulating tree defenses from I. sexdentatus (Levieux et al., 1991). However, in this later those rather specialized in invading the sapwood after tree de- species, no mycangial fungus has been identified and fungi- fenses are exhausted. In these cases, we present known ex- free insects develop the same way as fungi-contaminated bee- amples when fungus roles have been observed, or predictions tles (Colineau and Lieutier, 1984). Klepzig and Six (2004) when fungus roles are unknown. have suggested that, on poor substrates like dead trees or trees that die rapidly after attack, beetles need complementary food (mainly nitrogen and sterols) and could get it from associated fungi even if they are not mycangial. Possibly and similarly to 5.1. Group 1 – Beetle species not using the strategy the mycangial fungi, non mycangial fungi that serve as addi- of exhausting tree defenses and without an effective tional food may also serve as antagonists to blue stain fungi association with Ophiostomatoid fungi after tree resistance is overcome. This group gathers beetle species from both ends of the aggressiveness scale. 4.4. Multiple roles of associated fungal complexes, – Group 1a is composed of highly aggressive beetles at- a final hypothesis tacking living trees and succeeding in establishing and developing in their hosts without killing them, by using the strategy As a consequence of the discordance between fungal of avoiding tree defenses. They are not associated with fungi pathogenicity and beetle aggressiveness, of the tree killing because they must avoid stimulating tree defenses. Typical ex- process proposed above, and of our above comments and hy- amples are D. micans and D. punctatus on spruces. D. micans potheses on relationships between beetles and fungi, we fi- carries O. canum but at a very highly variable frequency (0.5 nally hypothesize that a beetle species using the strategy of to 90 %) and the fungus does not play any role in the success of overcoming tree resistance is generally associated with a fun- beetle establishment (Lieutier et al., 1992), hence correspond- gal complex, of which species could assume four roles re- ing to no real association. garding relations with beetles and trees: (1) to stimulate tree – Group 1b gathers the very secondary beetle species that defenses in the phloem and superficial sapwood, (2) to grow are able to attack living trees but with very low or with- into the sapwood after tree resistance is overcome, (3) to con- out defense ability. They do not need to exhaust tree de- trol phloem extension of the first other two categories, and fenses and thus have not developed fungal associations for es- (4) to bring nutrients to the beetle progeny, at least for the tablishing in trees. Some secondary beetles, that are weakly phloeomycetophagous beetles. Depending on beetle species or even frequently associated with more or less pathogenic and host trees, as well as environmental factors and beetle black stain fungi, also belong to this group. They are how- population levels, the nature and the relative abundance of the ever passive vectors, apparently unaffected by the presence beetle associated fungal species can vary (see Sects. 3.6 and 6, of the fungus, and their associations are rather commensal references included), and thus the above roles could be played (Klepzig and Six, 2004). For example, D. terebrans attacks by different fungal species, or the same fungal species could only very weak pines, although it has already been found to be involved in several roles, which could have consequences be highly associated (up to 100% of the beetles) with L. tere- for beetle population dynamics (Hofstetter et al., 2006). brantis, a fungus which, in addition to its high pathogenicity As suggested by Paine et al. (1997), multiple and complex (Harrington and Cobb, 1983), is able to induce violent tree re- interactions among beetle-associated fungi and between fungi actions (Rane and Tattar, 1987; Wingfield, 1983). Likewise,

801p15 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al. occasionally, Hylastes and Hylurgus carry L. serpens, L. pro- low density inoculations, and making predictions in situations cerum or L. wageneri (Harrington, 1993a; Jacobs and Wing- where this ability is still unknown. field, 2001; Klepzig et al., 1991). Hylurgops are also vectors In Table II, considering species where data on fungus abil- of L. procerum,andD. valens can carry L. serpens and even L. ity to stimulate tree defenses is available, the highly aggressive terebrantis (Klepzig et al., 1995; Jacobs and Wingfield, 2001). beetle species, D. frontalis, D. ponderosae, D. jeffreyi, D. pseu- In all these situations however, the insect doesn’t need fungus dotsugae, S. ventralis and I. typographus (Furniss and Carolin, to be present in order to establish itself on trees and certainly 1977; Grégoire and Evans, 2004) are all frequently associated only plays the role of a vector for the pathogen, making the with Ophiostomatoid fungi amongst which at least one species association not a true one (Klepzig et al., 1991). The fact that is able to highly stimulate tree phloem defenses (references in fungus has no role to play in beetle establishment on trees, Tab. II). This is the same for I. cembrae in Scotland, its area however, does not discard the possibility that associations with of introduction (Redfern et al., 1987). inversely, I. acuminatus, fungi have been developed to bring nutrients to the insect, at I. sexdentatus, O. erosus and I. pini, known to be moderately least at the larval stage (Eckhardt et al., 2004; Klepzig and Six, or weakly aggressive (Chararas, 1962; Furniss and Carolin, 2004). 1977; Grégoire and Evans, 2004), are all associated with fungal species that stimulate the phloem response only moderately or weakly (references in Tab. II). There is thus a close 5.2. Group 2 – Beetles species using the strategy correspondence between beetle species aggressiveness and the of exhausting tree defenses but very loosely ability of their main associated fungal species to stimulate tree associated with fungi defenses. The stimulating ability of the fungi associated with the Beetle species of this group attack living trees but only a other beetle species presented in Table II is not known but very low proportion of individuals carry fungi and/or the fungi we can make the following predictions. Since D. brevicomis are unable to stimulate tree defenses when introduced into the is known to be a very aggressive beetle for pines (Furniss tree by a beetle. The strategy of exhausting tree defenses is and Carolin, 1977), O. brevicomi certainly stimulates pine de- used but with little success because the fungal association is fenses very strongly during attacks by D. brevicomis. Simi- not effective enough to stimulate tree defenses. The best exam- larly, D. rufipennis is a highly aggressive beetle species on ple seems to be T. piniperda on pines during the reproduction spruce (Dumerski et al., 2001) and is highly and frequently phase of its life cycle when it attacks trunks. It is associated associated with L. abietinum (Aukema et al., 2005; Reynolds, with two blue stain fungi, L. wingfieldii rather pathogenic and 1992; Six and Bentz, 2003). We predict that L. abietinum O. minus moderately pathogenic, but the latter species has a strongly stimulates spruce defenses when introduced by D. ru- very low and variable frequency, while the former is unable fipennis in its host. Conversely, the aggressiveness of I. cem- to stimulate tree defenses when introduced into the tree by brae for the European larch is usually only moderate in con- a beetle (see above Sect. 3.6) (Lieutier et al., 1989a; 1989b, tinental Europe (Grégoire and Evans, 2004; Redfern et al., 1995). Probably Cryphalus abietis on fir (Kirschner, 1998 1987), its area of origin. Certainly, C. laricicola or another in Kirisits, 2004)andPityogenes quadridens on Scots pine highly associated fungal species stimulates larch defenses only (Mathiesen-Käärik, 1953; in Kirisits, 2004) also belong to this moderately in that area. group because of the low frequency of their associated fungi. At this stage of the discussion, one may wonder if a relation In this group, stimulation of tree defenses is only due to beetle exists between beetle aggressiveness and the ability of the fun- tunneling activity resulting, as observed during T. piniperda gal complex or other organisms to control blue stain extension reproductive attacks, in a high critical threshold of attack den- in the phloem. Indeed, when a possibility of controlling fun- sity (Långström and Hellqvist, 1993; Långström et al., 1992) gus extension exists, it should allow beetles to tolerate fungal and a very moderate or low aggressiveness (Långström and species which are particularly able to grow well in the phloem Hellqvist, 1988; Lieutier et al., 1995). and thus be able to strongly stimulate tree reactions. However the lack of sufficient knowledge on the existence of antagonistic species does not allow the validity of this speculation to be 5.3. Group 3 – Beetle species using the strategy checked. of exhausting tree defenses and highly associated with blue stain fungi 6. SPECULATIVE CONCLUSION Beetle species attack living trees, with a high percentage of individuals (sometimes up to 100%) carrying blue stain fungi. Bark beetles and fungi maintain very complex and multiple Within this group it is possible to distinguish beetle species interactions, making it difficult to propose a general model of according to their level of aggressiveness, while comparing it beetle - fungus relationships. Bark beetles vary considerably with the ability of their associated blue stain fungi to stimu- in their dependence on fungal complexes for nutrition and es- late tree defenses (Tab. II). The problem is however that, in tablishment on trees, according to both their strategy of estab- no situation, is the ability of the fungus to stimulate tree de- lishment and the physiological state of their usual host tree. fenses after introduction into the host tree by a beetle known. For the species using the strategy of exhausting tree defenses, We are thus using the fungus stimulating ability after artificial however, to be associated at a minimum frequency with fungi

801p16 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Table II. Aggressiveness of bark beetles species highly associated with fungi, and the ability of their main associated fungal species to stimulate host tree defenses. H = high; M = moderate; L = low.

Observation/ Beetle species Beetle Main fungal species stimulating tree defenses after low References / ability to prediction and host tree aggressiveness density inoculations, and level of stimulation (H, M, L) stimulate tree defenses D. frontalis H O. minus (H) 2, 3, 10, 15 (pines) D. ponderosae H G. clavigera (H), O. longiclavatum (H), O. montium (M) 13, 14, 17 Observation (pines) D. jeffreyi H G. clavigera (H) 17 (Jeffrey pine) D. pseudotsugae H O. europhioides (H), 18 (Douglas fir) O. pseudotsugae (H) S. ventralis H A. symbioticum (H) 19 (Grant fir) I. typographus H O. penicillatum (H), C. polonica (H) 4, 5, 16, (Norway spruce) Ambrosiella sp. (H) I. cembrae in UK H C. laricicola (H) 12 (introduction area) (European larch) D. brevicomis H O. brevicomi (H) Prediction (pines) D. rufipennis (spruces) H L. abietinum (H) I. acuminatus M O. ips (M), O. brunneo-ciliatum (M) 8 (pines) Observation I. sexdentatus M O. ips (M), O. brunneo-ciliatum (M) 6, 7 (pines) O. erosus M O.ips (M) 1, 9 (pines) I. pini M O. ips (M), O. nigrocarpum (L) 11 (pines) Prediction I. cembrae in cont. M C. laricicola (M) Europe (area of origin) or another species (M) (European larch) References regarding fungus ability to stimulate tree defenses: 1 = Ben Jamaa et al. (2007); 2 = Cook et al. (1986); 3 = Cook and Hain (1987); 4 = Krokene and Solheim (1997); 5 = Krokene and Solheim (1999); 6 = Lieutier et al. (1989b;7= Lieutier et al. (1990); 8 = Lieutier et al. (1991b); 9 = Lieutier et al. (2005); 10 = Paine and Stephen (1987); 11 = Raffa and Smalley (1988); 12 = Redfern et al. (1987); 13 = Rice et al. (2007a); 14 = Rice et al. (2007b); 15= Ross et al. (1992); 16 = Solheim (1988); 17 = Solheim and Krokene (1998a); 18 = Solheim and Krokene (1998b); 19 = Wong and Berryman (1977). able to naturally stimulate tree defenses is a necessity for an been to develop novel associations with additional fungi (or efficient attack. The association, even loosely, with Ophios- other micro-organisms) able to control the phloem extension tomatoid blue stain fungi allows that. of the former after defense exhaustion, thus allowing the de- For the aggressive beetle species, one may speculate that velopment of associations with fungi that were fast growing in no association existed at the origin, except in some cases for the phloem and very efficient in stimulating tree defenses. As nutrition, which is currently the case for the saprophytic and already mentioned, the most active sapwood invading species very secondary beetles of group 1b. The conquest of living are most often not those that have helped the beetle in previ- trees would have then diverged in two directions, one towards ously exhausting tree defenses. In such a context, most sap- avoiding tree defenses without developing a fungal associa- wood invading fungi might simply be “cheaters” which have tion (today group 1a), the other towards exploiting tree de- taken advantage of the increasing efficiency of the relation- fenses, with the progressive help of fungi. In this latter sit- ships between the beetles and the fungi stimulating tree de- uation, the association would have begun first as occasional, fenses. The rather high level of pathogenicity of some of them possibly comparable to the current situation for T. piniperda would result only from competition with other sapwood in- in group 2. Then, fungi would have been selected for their vaders. Such a scenario of building beetle-fungi associations ability to grow in the phloem and thus to stimulate tree de- based on primary association with fungal species specialized fenses (group 3). But this step would not allow selecting fungi in stimulating tree defense, completed later with the arrival of that were particularly efficient in stimulation because such ef- pathogenic “cheaters”, can be interestingly compared with the ficient fungi would grow too fast in the phloem after tree de- suggestion that weakly pathogenic fungi are “established as- fenses had been exhausted. The final step would thus have sociates” whereas pathogenic fungi are incidental late arriving

801p17 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

“invaders” (Six, 2003; Six and Paine, 1999; Six and Klepzig, Bridges J.R., 1983. Mycangial fungi of Dendroctonus frontalis 2004). (Coleoptera: Scolytidae) and their relationship to beetle population Moving from group 1b to groups 2 and then 3 gives the trends. Environ. Entomol. 12: 858–861. appearance of moving from very simple relationships (nutri- Bridges J.R., Nettleton W.A., and Conner M.D., 1985. Southern pine bee- tional only) to more and more elaborate relationships (nutri- tle (Coleoptera: Scolytidae) infestations without the blue-stain fun- tional plus stimulation plus eventual control of extension) be- gus, Ceratocystis minor. J. Econ. Entomol. 78: 325–27. tween beetles and fungal complexes. However, these succes- Brignolas F., Lieutier F., Sauvard D., Yart A., Drouet A., and Claudot sive groups do not reflect the phylogeny of bark beetles, at A.-C., 1995. Changes in soluble phenol content of Norway spruce (Picea abies Karst.) phloem in response to wounding and inoculation least when referring to the Dendroctonus genus (Kelley and with Ophiostoma polonicum. Eur. J. For. Pathol. 25: 253–65. Farrell, 1998; Six and Klepzig, 2004). The above speculated process of building beetle-fungi associations would have thus Brignolas F., Lieutier F., Sauvard D., Christiansen E., and Berryman ff A.A., 1998. Phenolic predictors for Norway spruce resistance to the evolved independently for the di erent beetle species of a bark beetle Ips typographus (Coleoptera: Scolytidae) and an associ- same group. This is in agreement with the conclusion that no ated fungus, Ceratocystis polonica. Can. J. For. Res., 28: 720–28. coevolutionary process has occurred between beetles and their Cardoza Y.J., Klepzig K.D., and Raffa K.F., 2006. Bacteria in oral se- associated fungi (Harrington, 1993a). cretions of an endophytic insect inhibit antagonistic fungi. Ecol. Entomol. 31: 636–645. Acknowledgements: We thank two anonymous reviewers for their Cardoza Y.J., Moser J.C., Klepzig K.D., and Raffa K.F., 2008 Multipartite very constructive and stimulating remarks on an earlier version of symbioses among fungi, mites, nematodes, and the spruce beetle, this paper. We also thank Caroline Sarré (University of Orleans) for Dendroctonus rufipennis. Environ. Entomol. 37: 956–963. proofreading the paper. Chararas C., 1962. Scolytides des conifères. Lechevalier Ed., Paris, 556p. Christiansen E., 1985a. Ceratocystis polonica inoculated in Norway REFERENCES spruce: Blue-staining in relation to inoculum density, resinosis and tree growth. Eur. J. For. Pathol. 15: 160–167. Anderbrandt O., Schlyter F., and Lofqvist J., 1988. Dynamics of tree attack in the bark beetle Ips typographus under semi-epidemic con- Christiansen E., 1985b. Ips/Ceratocystis infection of Norway spruce: ditions. In: T.L. Payne and H. Saarenmaa (Eds.), Integrated control what is a deadly dosage? Z. Angew. Entomol. 99: 6–11. of Scolytid bark beetle, Blacksburg: Virginia Polytech Institute State Christiansen E. and Glosli A.M., 1996. Mild drought enhances the resis- University, pp. 35–51. tance of Norway spruce to a bark beetle-transmitted blue-stain fun- Aukema B.H., Werner R.A., Haberkern K.E., Illman B.L., Clayton M.K., gus. In: W.J. Mattson, P. Niemela, and M. Rousi (Eds.), Dynamics and Raffa K.F., 2005. Quantifying sources of variation in the fre- of forest herbivory: Quest for pattern and principle, USDA For. Serv. quency of fungi associated with spruce beetles: Implications for hy- Gen. Tech. Rep. NC-183, pp. 192–99. pothesis testing and sampling methodology in bark beetle–symbiont relationships. For. Ecol. Manage., 10, 187–202. Christiansen E., Waring R.H., and Berryman A.A., 1987. Resistance of conifers to bark beetle attacks: searching for general relationships. Ayres M.P., Wilkens R.T., Ruel J.J., Lombardero M.J., and Vallery E., For. Ecol. Manage. 22: 89–106. 2000. Nitrogen budgets of phloem-feeding bark beetles with and without symbiotic fungi (Coleoptera: Scolytidae). Ecology 81: 2198– Colineau B. and Lieutier F., 1994. Production of Ophiostoma-free adults 2210. of Ips sexdentatus Boern. (Coleoptera: Scolytidae) and comparison Barras S.J., 1970. Antagonism between Dendroctonus frontalis and the with naturally contaminated adults. Can. Entomol. 126: 103–10. fungus Ceratocystis minor. Ann. Entomol. Soc. Am. 63: 1187–1190. Cook S.P. and Hain F.P., 1987. Four parameters of the wound response Basham H.G., 1970. Wilt of loblolly pine inoculated with blue-stain fungi of loblolly and shortleaf pines to inoculations with the blue-staining of the genus Ceratocystis. Phytopathology 60: 750–754. fungus associated with the southern pine beetle. Can. J. Bot. 65: 2403–2409. Ben Jamaa M.L., Lieutier F., Yart A., Jerraya A., and Khouja M.L., 2007. Phytopathogenic fungi associated with the bark beetles Tomicus Cook S.P., Hain F.P., and Nappen P.B., 1986. Seasonality of hypersen- piniperda and Orthotomicus erosus in Tunisia; study of the virulence sitive response by loblolly and shortleaf pine to inoculation with a of some isolates. For. Pathol. 37: 51–63. fungal associate of the southern pine beetle (Coleoptera: Scolytidae). Berryman A.A., 1972. Resistance of conifers to invasion by bark beetle- J. Entomol. Sci. 21: 283–285. fungus associations. BioScience 22: 298–602. Coulson R.N., 1979, Poulation dynamics of bark beetles. Annu. Rev. Berryman A.A., 1976. Theoretical explanation of mountain pine betle dy- Entomol. 24: 417-447. namics in lodgepole pine forests. Environ. Entomol. 5: 1225–1233. Croisé L. and Lieutier F., 1993, Effect of drought on the induced defence Berryman A.A., 1982. Population dynamics of bark beetles. In: J.B. reaction of Scots pine to bark beetle associated fungi. Ann. Sci. for. Mitton and K.B. Sturgeon (Eds.), Bark beetles in North American 50: 91–97. conifers, Austin, Univ. Texas, pp. 264–314. Croisé L., Lieutier F., and Dreyer E., 1998a. Scots pine responses to num- Bleiker K.P. and Six D.L., 2007. Dietary benefits of fungal associates to ber and density of inoculation points with Leptographium wingfieldii an eruptive herbivore: potential implications of multiple associates Morelet, a bark beetle-associated fungus. Ann. Sci. For. 55: 497–506. on host population dynamics. Environ. Entomol. 36: 1384–1396. ff Bleiker K.P. and Six D.L., 2009. Competition and coexistence in a multi- Croisé L., Dreyer E., and Lieutier F., 1998b, E ects of drought stress and partner mutualism: interactions between two fungal symbionts of severe prunning on the reaction zone induced by single inoculations the mountain pine beetle in beetle-attacked trees. Microb. Ecol. 57: with a bark beetle associated fungus (Ophiostoma ips) in the phloem 191–202. of young Scots pines. Can. J. For. Res. 28: 1814–1824. Bois E. and Lieutier F. 1997. Phenolic response of Scots pine clones to Delorme L. and Lieutier F., 1990. Monoterpene composition of the pre- inoculation with Leptographium wingfieldii, a fungus associated with formed and induced resins of Scots pine, and their effect on bark Tomicus piniperda. Plant Physiol.Biochem. 35: 819–825. beetles and associated fungi. Eur. J. For. Pathol. 20: 304–16.

801p18 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Dymerski A.D., Anhold J.A., and Munson A.S., 2001. Spruce beetle and commensalistic interactions among insects, fungi and mites. (Dendroctonus rufipennis) outbreaks in Engelmann spruce (Picea en- Community Ecol. 8: 47–56. gelmanni) in central Utah, 1986–1998. West. N. Am. Nat., 61: 19–24. Horntvedt R., Christiansen E., Solheim H., and Wang, S., 1983. Artificial Eckhardt L.G., Goyer R.A., Klepzig K.D., and Jones J.P., 2004. inoculation with Ips typographus-associated blue-stain fungi can kill Interactions of Hylastes species (Coleoptera: Scolytidae) with healthy Norway spruce trees. Medd. fra Norsk. Skogforsk. 38: 1–20. Leptographium species associated with loblolly pine decline. J. Econ. Leptographium Entomol. 97: 468–474. Jacobs K. and Wingfield M.J., 2001. species, Insect associates and Agents of Blue Stain, APS Press, St Paul, Minnesota, Filip G.M., Christiansen E., and Parks C., 1989. Secondary resin pro- 207 p. duction increases with vigor of Abies grandis inoculated with Trichosporium symbioticum in Northeastern Oregon. USDA FS, Jankowiak R., 2005. Fungi associated with Ips typographus on Picea PNW-RN 489, 1–10. abies in southern Poland and their succession into the phloem and sapwood of beetle-infested trees and logs. For. Pathol. 35: 37–55. Franceschi V.R., Krokene P., Christiansen E., and Krekling T., 2005. Anatomical and chemical defenses of conifers against bark beetles Kelley S.T. and Farrell B.D., 1998. Is specialization a dead end? The and other pests. New Phytol. 167: 353–376. phylogeny and host use in Dendroctonus bark beetles (Scolytidae). Evolution 52: 1731–1743. Francke-Grosmann H., 1963. Some new aspects in forest entomology. Annu. Rev. Entomol. 8: 415–438. Kim J-J., Plattner A., Lim Y.W., and Breuil C., 2008. Comparison of two Francke-Grosmann H., 1967. Ectosymbiosis in wood-inhabiting insects. methods to asses the virulence of the mountain pine beetle associate, In: S.M. Henry (Ed.), Symbiosis, Vol. 2, Academic Press, New York, Grosmannia clavigera,toPinus contorta. Scand. J. For. Res. 23: 98– London. 104. Furniss R.L. and Carolin V.M., 1977. Western Forest Insects. USDA For. Kirisits T., 1998. Pathogenicity of three blue-stain fungi associated with Serv. Misc. Publ. N◦ 1339, 654 p. the bark beetle Ips typographus to Norway spruce in Austria. Österr. Z. Pilzk. 7: 191–201. Grégoire J.C. and Evans H.F., 2004. Damage and control of BAWBILT organisms, an overview. In: F. Lieutier, K.R. Day, A. Battisti, J.C. Kirisits T., 1999. Report on a strain of the pathogenic blue-stain fungus Grégoire, and H.J. Evans (Eds.), Bark and wood boring insects in Ceratocystis polonica with low virulence. Osterr. Z. Piltz. 8: 157– living trees in Europe, a synthesis, Kluwer Acad. Publ., Dordrecht, 167. pp. 19–37. Kirisits T., 2001. Studies on the association of ophiostomatoid fungi Guérard N., Dreyer E., and Lieutier F., 2000. Interactions between with bark beetles in Austria with special emphasis on Ips ty- Scots pine, Ips acuminatus (Gyll.) and Ophiostoma brunneo-ciliatum pographus and Ips cembrae and their associated fungi Ceratocystis (Math.): estimation of critical thresholds of attack and inoculation polonica and Ceratocystis laricicola. Dissertation thesis, Universität densities and effect on hydraulic properties of the stem. Ann. For. für Bodenkultur Wien. Sci. 57: 681–690. Kirisits T., 2004. Fungal associates of European bark beetles with special Harding S., 1989. The influence of mutualistic blue-stain fungi on emphasis on the Ophiostomatoid fungi. In: F. Lieutier, K.R. Day, A. bark beetle population dynamics. Ph.D thesis, Royal Veterinary and Battisti, J.C. Grégoire, and H.F. Evans (Eds.), Bark and wood boring Agricultural University Copenhagen. insects in living trees in Europe, a synthesis, Kluwer, Dordrecht, pp. Harrington T.C., 1993a. Diseases of conifers caused by species of 181–235. Ophiostoma and Leptographium. In: M.J. Wingfield, K.A. Seifert, Kirisits T., Grubelnik R., and Führer E., 2000. Die ökologis- and J.F. Webber (Eds.), Ophiostoma and Ceratocystis: Taxonomy, che Bedeutung von Bläuepilzen für rindenbrütende Borkenkäfer. Ecology and Pathogenicity, APS Press, St Paul, Minnesota, pp. 161– In: F. Müller (Ed.), Mariabrunner Waldbautage 1999 – Umbau 172. sekundärer Nadelwälder, Vienna, Schriftenreihe der Forstlichen Harrington T.C., 1993b. Biology and taxonomy of fungi associated with Bundesversuchsanstalt Wien, FBVA-Berichte, 111, pp. 117–137. bark beetles. In: T.D. Schowalter and G.M. Filip (Eds.), Beetle- Klepzig K.D. and Six D.L., 2004. Bark beetle-fungal symbiosis: context Pathogen Interactions in Conifer Forests, New York: Academic dependency in complex associations. Symbiosis 37: 189–205. Press, pp. 37–58. Klepzig K.D., Raffa K.F., and Smalley E.B., 1991. Association of an Harrington T.C., 2005. Ecology and evolution of mycetophagous bark insect-fungal complex with red pine decline in Wisconsin. For. Sci. beetles and their fungal partners. In: F.E. Vega and M. Blackwell 37: 1119–1139. (Eds.), Insect-fungal associations, ecology and evolution, Oxford University Press, pp. 257–289. Klepzig K.D., Smalley E.B., and Raffa K.F., 1995. Dendroctonus Harrington T.C. and Cobb F.W. Jr., 1983. Pathogenicity of Leptographium valens and Hylastes porculus (Coleoptera: Scolytidae): vectors of and Verticicladiella spp. Isolated from roots of western North pathogenic fungi (Ophiostomatales) associated with red pine decline American conifers. Phytopathology 73: 596–599. disease. Great Lakes Entomol. 28: 81–87. Harrington T.C. and Wingfield M.J., 1998. The Ceratocystis species on Klepzig K.D., Moser J.C., Lombardero M.J., Ayres M.P., Hofstetter R.W., conifers. Can. J. Bot. 76: 1446–57 and Walkinshaw C.J., 2001a. Mutualism and antagonism: ecological interactions among bark beetles, mites and fungi. In: M.J. Jeger Hetrick L.A., 1949. Some overlooked relationships of southern pine bee- and N.J. Spence (Eds.), Biotic interactions in plant-pathogen associ- tle. J. Econ. Entomol. 42: 466–469. ations, CAB International, pp. 237–267. Highley L. and Tattar T.A., 1985. Leptographium terebrantis and black Klepzig K.D., Moser J.C., Lombardero F.J., Hofstetter R.W., and Ayres turpentine beetles associated with blue stain and mortality of black M.P., 2001b. Symbiosis and competition: complex interactions and Scots pine on Cape Cod, Massachussets. Plant Dis. 69: 528–530. among beetles, fungi and mites. Symbiosis 30: 83–96. Hobson K.R., Parmeter J.R. Jr., and Wood D.L. 1994. The role of Knizek M. and Beaver R., 2004. Taxonomy and systematic of bark and fungi vectored by Dendroctonus brevicomis Leconte (Coleoptera: ambrosia beetles. In: F. Lieutier, K.R. Day, A. Battisti, J.C. Grégoire, Scolytidae) in occlusion of ponderosa pine xylem. Can. Entomol. and H.J. Evans (Eds.), Bark and wood boring insects in living trees 126: 277–282. in Europe, a synthesis, Kluwer Acad. Publ., Dordrecht, pp. 41–54. Hofstetter R.W., Cronin J.T., Klepzig K.D., Moser J.C., and Ayres M.P., Krokene P. 1996. The role of blue-stain fungi in tree-killing by bark bee- 2006. Antagonisms, mutualisms and commensalisms affect outbreak tles. Thesis, University of Oslo. dynamics of the southern pine beetle. Oecologia, 147: 679–691. Krokene P. and Solheim H., 1997. Growth of four bark-beetle-associated Hofstetter R.W., Dempsey T.D., Klepzig K.D., and Ayres M.P., blue-stain fungi in relation to the induced wound response in Norway 2007. Temperature-dependent effects on mutualistic, antagonistic, spruce. Can. J. Bot. 75: 618–625.

801p19 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

Krokene P. and Solheim H., 1998. Pathogenicity of four blue-stain défense des pins sylvestres (Pinus sylvestris L.) inoculés. Agronomie fungi associated with aggressive and nonaggressive bark beetles. 10: 243–56. Phytopathology 88: 39–44 Lieutier F., Yart A., Jay-Allemand C., and Delorme L. 1991a. Preliminary Krokene P. and Solheim H., 1999. What do low density inoculations with investigations on phenolics as a response of Scots pine phloem to fungus tell us about fungal virulence and tree resistance? In: Lieutier attacks by bark beetles and associated fungi. Eur. J. For. Pathol. 21: F. Mattson W.J., and Wagner M.R. (Eds.), Physiology and genetics 354–354. of tree-phytophage interactions, INRA Editions, Versailles, pp. 353– Lieutier F., Garcia J., Yart A., Vouland G., Pettinetti M., and Morelet M., 362. 1991b. Ophiostomatales (Ascomycetes) associées à Ips acuminatus Långström B. and Hellqvist C., 1988. Scots pine resistance against (Coleoptera: Scolytidae) sur le pin sylvestre (Pinus sylvestris)dans Tomicus piniperda as related to tree vitality and attack density. In: le Sud-Est de la France, et comparaison avec Ips sexdentatus Boern. Payne T.L. and Saarenmaa H. (Eds.), Integrated control of scolytid Agronomie 11: 807–817. bark beetles, Virginia Polytech. Inst. St. Univ., Blacksburg, pp. 121– Lieutier F., Vouland G., Pettinetti M., Garcia J., Romary P., and Yart 133. A., 1992. Defence reactions of Norway spruce (Picea abies Karst.) to artificial insertion of Dendroctonus micans Kug. (Coleoptera: Långström B. and Hellqvist C., 1993. Scots pine susceptibility to attack Scolytidae). J. Appl. Entomol. 114: 174–186. by Tomicus piniperda (L) as related to pruning date and attack density. Ann. Sci. For. 50: 101–117. Lieutier F., Garcia J., Yart A., and Romary P., 1995. Wound reaction of Scots pine (Pinus sylvestris L.) to attacks by Tomicus piniperda Långström B., Hellqvist C., Ericsson A., and Gref D., 1992. Induced L. and Ips sexdentatus Boern. (Coleoptera: Scolytidae). J. Appl. defense reaction in Scots pine following stem attacks by Tomicus Entomol. 119: 591–600. piniperda L. Ecography 15: 318–327. Lieutier F., Sauvard D., Brignolas F., Picron V., Yart A., Bastien C., Långström B., Solheim H., Hellqvist C., and Gref R., 1993. Effects and Jay-Allemand C., 1996. Changes in phenolic metabolites of of pruning young Scots pines on host vigour and susceptibility to Scots pine phloem induced by Ophiostoma brunneo-ciliatum,abark- Leptographium wingfieldii and Ophiostoma minus, two blue-stain beetle-associated fungus. Eur. J. For. Pathol., 26, 145–158. fungi associated with Tomicus piniperda. Eur. J. For. Pathol. 23: 400– Lieutier F., Yart A., Ye H., Sauvard D., and Gallois V., 2004. Between- 415. isolate variations in the performances of Leptographium wingfieldii Lee S., Kim J.-J., and Breuil C., 2006. Diversity of fungi associated Morelet, a fungus associated with the bark beetle Tomicus piniperda with mountain pine beetle, Dendroctonus ponderosae,andinfested L. Ann. For. Sci. 61: 45–53. lodgepole pines in British Columbia. Mountain Pine beetle Research Lieutier F., Ghaioule D., and Yart A., 2005. Virulence and possible Initiative Working Paper 2006-06, 20 p. role of fungi associated with the bark beetles Tomicus piniperda L., and Orthotomicus erosus Woll. in Morocco. In: F. Lieutier and D. Lévieux J., Cassier P., Guillaumin D., and Roques A., 1991. Structures Ghaioule (Eds.), Entomological research in Mediterranean ecosys- implicated in the transportation of pathogenic fungi by the European tems, INRA Ed., Versailles, pp. 195–208. bark beetle, Ips sexdentatus Boerner: Ultrastructure of a mycangium. Can. Entomol. 123: 245–54. Mathiesen A., 1950. Über einige mit Borkenkäfern assoziierte Bläuepilze in Schweden. Oikos 2: 275–308. Lieutier F., 1992. Les réactions de défense des conifères et stratégies d’attaque de quelques Scolytides européens. Mém. Soc. Roy. belge Mathiesen-Käärik A., 1953. Eine Übersicht über die gewöhnlichsten mit Entomol. 35: 529–539. Borkenkäfern assoziierten Bläuepilze in Schweden und einige für Schweden neue Bläuepilze. Meddel. Frå. Stat. Skogforsk. Inst. 43: Lieutier F., 1995. Associated fungi, induced reaction and attack strat- 1–74. egy of Tomicus piniperda (Coleoptera: Scolytidae) in Scots pine. ff Mathre D.E., 1964a. Survey of Ceratocystis spp. associated with bark In: F.P. Hain, S.M. Salom, W.F. Ravlin, T.L. Payne, and K.F. Ra a beetles in California. Contrib. Boyce Thomp. Inst. 22: 353–361. (Eds.), Behavior, Population Dynamics and Control of Forest Insects, Proceedings International Union Forestry Research Organizations Mathre D.E., 1964b. Pathogenicity of Ceratocystis ips and Ceratocystis Joint Conference, February 1994 Maui, Hawaï, pp. 139–153. minor to Pinus ponderosa. Contrib. Boyce Thomp. Inst. 22: 363–388. Neal T.A. and Ross D.W., 1999. Pathogenicity to western larch Lieutier F., 2002. Mechanisms of resistance in conifers and bark beetle (Larix occidentalis) of two fungi, Ophiostoma pseudotsugae and attack strategies. In: M.R. Wagner, K.M. Clancy, F. Lieutier, and T.D. Leptographium abietinum, associated with the Douglas fir beetle Paine (Eds.), Mechanisms and deployment of resistance in trees to (Coleoptera: Scolytidae). Agric. For. Entomol. 1: 203–207. insects, Kluwer, Dordrecht, pp. 31–75. Nebeker T.E., Hodges J.D., and Blanche C.A. 1993. Host Response to Lieutier F., 2004. Host resistance to bark beetles and its variations. In: Bark Beetle and Pathogen Colonization. In. T.D. Schowalter and F. Lieutier, K.Day, A. Battisti, J.C. Grégoire, and H. Evans (Eds.), G.M. Filip (Eds.), Beetle Pathogen Interactions in Conifer Forests, Bark and wood boring insects in living trees in Europe, A synthesis, Academic Press, San Diego, pp. 157–173. Kluwer, Dordrecht, pp. 135–180. Owen D.R., Lindahl K.Q., Wood D.L., and Parmeter J.R.Jr., 1987. Lieutier F., Yart A., Garcia J., Poupinel B., and Lévieux J., 1988. Do Pathogenicity of fungi isolated from Dendroctonus valens, D. fungi influence the establishment of bark beetles in Scots pine? In: brevicomis and D. ponderosae to ponderosa pine seedlings. W.J. Mattson, J. Lévieux, and C. Bernard-Dagan (Eds.), Mechanisms Phytopathology 77: 631–636. of woody plant defenses against insects: search for pattern, Springer, New York, pp. 321–334. Paine T.D. and Birch M.C., 1983. Acquisition and maintenance of mycangial fungi by Dendroctonus brevicomis LeConte (Coleoptera: Lieutier F., Cheniclet C., and Garcia J., 1989a. Comparison of the de- Scolytidae). Environ. Entomol. 12: 1384–1386. fense reactions of Pinus pinaster and Pinus sylvestris to attacks by two bark beetles (Coleoptera: Scolytidae) and their associated fungi. Paine T.D. and Stephen F.M., 1987. Fungi associated with the southern Environ. Entomol. 18: 228–234. pine beetle: avoidance of induced defense response in loblolly pine. Oecologia 74: 377–379. Lieutier F., Yart A., Garcia J., Ham M.C., Morelet M., and Lévieux J., 1989b. Champignons phytopathogènes associés à deux Coléoptères Paine T.D., Raffa K.F., and Harrington T.C., 1997. Interactions among Scolytidae du pin sylvestre (Pinus sylvestris L.) et étude préliminaire Scolytid bark beetles, their associated fungi, and live host conifers. de leur agressivité envers l’hôte. Ann. Sci. For. 46: 201–216. Annu. Rev. Entomol. 42: 179–206. Lieutier F., Yart A., Garcia J., and Ham M-C., 1990. Cinétique Parmeter J.R., Slaughter G.W., Chen M.M., and Wood D.L., 1992. Rate de croissance des champignons associés à Ips sexdentatus Boern. and depth of sapwood occlusion following inoculation of ponderosa et à Tomicus piniperda L. (Coleoptera: Scolytidae) et des réactions de pine with bluestain fungi. For. Sci. 38: 34–44.

801p20 Fungi and attack success of bark beetles Ann. For. Sci. 66 (2009) 801

Payne T.L., 1980. Life History & Habits. In: R.C. Thatcher, J.L. Searcy, Ross D.W. and Solheim H., 1997. Pathogenicity to Douglas fir of J.E. Coster, and G.D. Hertel (Eds.), The southern pine beetle, USDA Ophiostoma pseudotsugae and Leptographium abietinum, fungi as- For. Serv. Sci. Educ. Admin. Tech. Bull. N◦ 1631, pp. 7–28. sociated with the Douglas fir beetle. Can. J. For. Res. 27: 39–43. Plattner A., Kim J-J., DiGiustini S., and Breuil C., 2008. Variation Ross D.W., Fenn P., and Stephen F.M., 1992. Growth of southern pine in pathogenicity of a mountain pine beetle-associated fungus, beetle associated fungi in relation to the induced wound response in Grosmannia clavigera, on young lodgepole pine in British Columbia. loblolly pine. Can. J. for. Res. 22: 1851–1859. Can J. Plant Pathol. 30: 1–10. Russel C.E. and Berryman A.A., 1976. Host resistance to the fir en- Raffa K.F., 1991. Induced defensive reactions in conifer-bark beetle sys- graver. 1. Monoterpene composition of Abies grandis pitch blisters tems. In: D.W. Tallamy and M.J. Raupp (Eds.), Phytochemical in- and fungus-infested wounds. Can. J. Bot. 54: 14–18. duction by herbivore, Wiley and Sons, pp. 245–276. Safranyik L., Shrimpton D.M., and Whitney H.S., 1975. An Interpretation Raffa K.F., 2001. Mixed messages across multiple trophic levels: the ecol- of the Interaction Between Lodgepole Pine, the Mounain Pine Beetle ogy of bark beetle chemical communication systems. Chemoecology and its Associated Blue Stain Fungi in Western Canada. In: D. 11: 49-65. Baumgartner (Ed.), Management of Lodgepole Pine Ecosystems, Raffa K.F. and Berryman A.A., 1982a. Accumulation of monoterpenes Pullman: Wash. State Univ. Coop. Ext. Serv., pp. 406–428. and associated volatiles following fungal inoculation of grant fir Sallé A., Monclus R., Yart A., Garcia J., Romary P., and Lieutier F., 2005. with a fungus transmitted by the fir engraver Scolytus ventralis Fungal flora associated with Ips typographus: Frequency, virulence (Coleoptera: Scolytidae). Can. Entomol. 114: 797-810. and ability to stimulate the host defence reaction in relation to insect Raffa K.F. and Berryman A.A., 1982b. Gustatory cues in the orientation population levels. Can. J. For. Res. 35: 365–373. of Dendroctonus ponderosae (Coleoptera: Scolytidae) to host trees. Sallé A., Ye H., Yart A., and Lieutier F., 2008. Pinus yunnanensis resis- Can. Entomol. 114: 97-104. tance to a bark beetle associated fungus is increased during a natural Raffa K.F. and Berryman A.A., 1983a. The role of host plant resistance water stress. Tree Physiol. 28: 679–687. in the colonization behavior and ecology of bark beetles (Coleoptera: Siemaszko W., 1939. Zepoly grzybów towarzyszacych kornikom polskim Scolytidae). Ecol. Monog. 53: 27–49. [Fungi associated with bark-beetles in Poland]. Planta Polonica 7: 1– + Raffa K.F. and Berryman A.A., 1983b. Physiological aspects of lodge- 54 plates. pole pine wound responses to a fungal symbiont of the moun- Six D.L., 2003. Bark beetle-fungus symbiosis. In: T. Miller and K. tain pine beetle, Dendroctonus ponderosae (Coleoptera: Scolytidae). Kourtzis (Eds.), Insect symbiosis, CRC Press, pp. 97–114. Can. Entomol. 115: 723–34. Six D.L. and Bentz B.J., 2003. Fungi associated with the North American ff Ra a K.F. and Klepzig K.D., 1992. Tree defense mechanisms against spruce beetle, Dendroctonus rufipennis. Can. J. For. Res. 33: 1815– fungi associated with insects. In: R.A. Blanchette and A.R. Biggs 1820. (Eds.), Defense Mechanisms of Woody Plants Against Fungi, Springer, New York, pp. 354–389. Six D.L. and Bentz B.J., 2007. Temperature determines symbiont abundance in a multipartite bark beetle-fungus ectosymbiosis. Microb. ff Ra a K.F. and Smalley E.B., 1988. Response of red and Jack pines to Ecol. 54: 112–118. inoculation with microbial associates of the pine engraver, Ips pini (Coleoptera: Scolytidae). Can. J. For. Res. 18: 581–586. Six D.L. and Klepzig K.D., 2004. Dendroctonus bark beetles as model systems for studies on symbiosis. Symbiosis 37: 207–232. Raffa K.F., Phillips T.W., and Salom S.M., 1993. Strategies and mechanisms of host colonization by bark beetles. In: Schowalter T.D. and Six D.L. and Paine T.D., 1997. Ophiostoma clavigerum is the mycangial Filip G.M. (Eds.), Beetle-pathogen interactions in conifer forests, fungus of the Jeffrey pine beetle, Dendroctonus jeffreyi. Mycologia Academic Press, San Diego, pp. 103–128. 89: 858–866. Rane K.K. and Tattar T.A., 1987. Pathogenicity of blue stain fungi asso- Six D.L. and Paine T.D., 1998. Effects of mycangial fungi and host tree ciated with Dendroctonus terebrans. Plant Dis. 71: 879–883. species on progeny survival and emergence of Dendroctonus ponderosae (Coleoptera: Scolytidae). Environ. Entomol. 27: 1393–1401. Redfern D.B., Stoakley J.T., Steele H., and Minter D.W., 1987. Dieback and death of larch caused by Ceratocystis laricicola sp.nov.follow- Six D.L. and Paine T.D., 1999. Phylogenetic comparison of ascomycete ing attack by Ips cembrae. Plant Pathol. 36: 467–80. mycangial fungi and Dendroctonus bark beetles (Coleoptera: Scolytidae). Ann. Entomol. Soc. Am. 91: 159–166. Reid R.W., Whitney H.S., and Watson J.A., 1967. Reactions of lodgepole pine to attack by Dendroctonus ponderosae Hopkins and blue stain Solheim H., 1986. Species of Ophiostomataceae isolated from Picea fungi. Can. J. Bot. 45: 115–126. abies infested by the bark beetle Ips typographus. Nord. J. Bot. 6: 199–207 Rennerfelt E., 1950. Über den Zusammenhang zwischen dem Verblauen des Holzes und den Insekten. Oikos 2: 120–37 Solheim H., 1988. Pathogenicity of some Ips typographus associated blue-stain fungi to Norway spruce. Medd. fra Norsk. Inst. Skogforsk. Reynolds K.M., 1992. Relations between activity of Dendroctonus 40: 1–11. rufipennis Kirby on Lutz spruce and blue stain associate with Leptographium abietinum (Peck) Wingfield. For. Ecol. Manage. 47: Solheim H., 1991. Oxygen deficiency and spruce resin inhibition of 71–86. growth of fungi associated with Ips typographus. Mycol. Res. 95: 1387–1392. Rice A.V., Thormann M.N., and Langor D.W., 2007a. Mountain pine beetle associated blue-stain fungi cause lesions on jack pine, lodgepole Solheim H., 1992a. The early stages of fungal invasion in Norway spruce pine, and lodgepole x jack pine hybrids in Alberta. Can. J. Bot. 85: infested by the bark beetle Ips typographus. Can. J. Bot. 70: 1–5. 307–315. Solheim H., 1992b. Fungal succession in sapwood of Norway spruce in- Rice A.V., Thormann M.N., and Langor D.W., 2007b. Virulence of, and fested by the bark beetle Ips typographus. Eur. J. For. Pathol. 22: interactions among, mountain pine beetle associated blue-stain fungi 136–48. on two pine species and their hybrids in Alberta. Can. J. Bot. 85: Solheim H., 1993a. Ecological aspects of fungi associated with the spruce 316–323. bark beetle Ips typographus in Norway. In: Wingfield M.J., Seifert Rice A.V., Thormann M.N., and Langor D.W., 2008. Mountain pine K.A., and Webber J.F. (Eds.), Ceratocystis and Ophiostoma: taxon- beetle-associated blue-stain fungi are differentially adapted to boreal omy, ecology and pathogenicity, APS Press, St Paul, Minnesota, pp. temperatures. For. Pathol. 38: 113–123. 235–242.

801p21 Ann. For. Sci. 66 (2009) 801 F. Lieutier et al.

Solheim H., 1993b. Fungi associated with the spruce bark beetle Ips ty- Wallin K.F. and Raffa K.H., 2001. Effects of folivory on subcortical plant pographus in an endemic area in Norway. Scand. J. For. Res. 8: 118– defenses: can defense theories predict interguild processes? Ecology 122. 82: 1387–1400. Solheim H. and Krokene P., 1998a. Growth and virulence of mountain Waring R.H. and Pitman G.B., 1983. Physiological stress in lodgepole pine beetle associated blue-stain fungi, Ophiostoma clavigerum and pine as a precursor for mountain pine beetle attack. Z. Angew. Ophiostoma montium. Can. J. Bot. 76: 561–566. Entomol. 96: 265–270. Solheim H. and Krokene P., 1998b. Growth and virulence of Ceratocystis rufipenni and three blue-stain fungi isolated from the Douglas-fir Whitney H.S., 1982. Relationships between bark beetles and symbiotic beetle. Can. J. Bot. 76: 1763–1769. organisms. In: Mitton J.B. and Sturgeon K.B. (Eds.), Bark beetles in Solheim H. and Långström B., 1991. Blue stain fungi associated with North American conifers, Austin, Univ. Texas Press, pp. 183–211. Tomicus piniperda in Sweden and preliminary observation on their Whitney H.S. and Cobb F.W., 1972. Non-staining fungi associated with pathogenicity. Ann. Sci. For. 48: 149–156. the bark beetle Dendroctonus brevicomis (Coleoptera: Scolytidae) on Solheim H. and Safranyik L., 1997. Pathogenicity to Sitka spruce Pinus ponderosa. Can. J. Bot. 50: 1943–1945. of Ceratocystis rufipenni and Leptographium abietinum, blue-stain fungi associated with the spruce bark beetle. Can. J. For. Res. 27: Whitney H.S. and Farris S.H., 1970. Maxillary mycangium in the moun- 1336–1341. tain pine beetle. Science 167: 54–55. Solheim H., Långström B., and Hellqvist C., 1993. Pathogenicity of the Wingfield M.J., 1983. Association of Verticicladiella procera and blue stain fungi Leptographium wingfieldii and Ophiostoma minus to Leptographium terebrantis with insects in the lake states. Can. J. For. Scots pine: effect of tree pruning and inoculum density. Can J. For. Res. 13: 1238–1245. Res. 23: 1438–1443. ff Wingfield M.J., Seifert K.A., and Webber J.F., 1993. Ceratocystis and Solheim H., Krokene P., and Långström B., 2001. E ects of growth Ophiostoma: taxonomy, ecology, and pathogenicity, APS Press, and virulence of associated blue-stain fungi on host colonization be- 293 p. haviour of the pine shoot beetles Tomicus minor and T. piniperda. Plant Pathol. 50: 111–16. Wong B.E. and Berryman A.A., 1977. Host resistance to the fir engraver Stephen F.M., Berisford C.W., Dahlsten D.L., Fenn P., and Moser beetle. 3. Lesion development and containment of infection by resis- J.C., 1993. Invertebrate and microbial associates. In: Schowalter T. tant Abies grandis inoculated with Trichosporium symbioticum.Can. and Filip G. (Eds.), Beetle-pathogen interactions in conifer forests, J. Bot. 55: 2358–2365. Academic, San Diego, pp. 129–153. Wood D.L., 1982. The role of pheromones, kairomones, and allomones in Uzunovic A. and Webber J.F., 1998. Comparison of blue stain fungi the host selection and behavior of bark beetles. Annu. Rev. Entomol. grown in vitro and in freshly cut pine billets. Eur. J. For. Pathol. 28: 27: 411–446. 323–334. Viiri H., 1997. Fungal associates of the spruce bark beetle Ips typogra- Wullschleger S.D., McLaughlin S.B., and Ayres M.P., 2004. High reso- phus L. (Col., Scolytidae) in relation to different trapping methods. lution analysis of stem increment and sap flow for loblolly pine trees J. Appl. Entomol. 121: 529–33. attacked by southern pine beetle. Can. J. For. Res. 34: 2387–2393. Viiri H. and Lieutier F., 2004. Ophiostomatoid fungi associated with Yamaoka Y., Hiratsuka Y., and Maruyama P.J., 1995. The ability of the spruce bark beetle, Ips typographus, in post-epidemic areas in Ophiostoma clavigerum to kill mature lodgepole pine trees. Eur. J. France. Ann. For. Sci. 61: 215–219. For. Pathol. 25: 401–404.

801p22